When I was a kid, you heard “Good job” quite a lot. Successful double in little league? “Good job!” 93% on a math test? “Good job!” In contrast, the phrase “Good effort” was reserved for … lesser things. You struck out but … you tried hard. That’s a good effort, but not a good job.
The world of child praise has, however, shifted dramatically. Saying “Good job” has approached taboo status. The social media accounts of parenting experts are rife with advice to avoid this type of praise, in all its forms — no “Amazing!” or “You’re so smart.” Instead, we’re encouraged to applaud effort, not achievement: “It’s great to hear you worked hard on that.” Often, parents are told that it’s better to say nothing at all.
This advice is well-meant, but it can become yet another for parents to feel like we’re failing. It can also be paralyzing. A couple of months ago, my daughter told me how she did on a math test. Worried about saying the wrong thing, I just said “OK.” Which didn’t feel quite right either.
The underlying reason for the praise shift is, more or less, based on data. Perhaps the most famous and widely cited paper is “Praise for intelligence can undermine children's motivation and performance” published in the Journal of Personality and Social Psychology in 1998. In this paper, Carol Dweck and Claudia Mueller report the results of a number of experiments with fifth graders in which students attempted various tasks and were praised for either their intelligence or for their effort. In general, they found that those who were praised for their effort were more interested in pursuing harder problems, and more likely to feel they could improve.
Based on this and related research, Dweck introduced the world to the idea of the “growth mindset.” It’s broader than this one element, but a key aspect is the idea of focusing on kids’ efforts, rather than their ability.
This research is interesting and compelling. It makes a strong case for encouraging a growth mindset in school and helping children see the value of perseverance. What it doesn’t do — at least not directly — is suggest that you should never tell your child “Good job!” That leap — from interesting research to parenting polemic — that is a leap that the parenting-industrial complex has made all on our own.
The parenting industrial complex has a long track record of this type of overreaction. Think about the advice to talk to your baby all the time. It stems largely from the work of two academics in the mid-1990s. They worked with 72 families in Kansas across the income spectrum and found that the number of words children heard by age 3 differed widely — by perhaps 30 million words — across the socioeconomic spectrum. They, and others, argued that this exposure to language was key to academic and social development.
These are extremely interesting findings, and they may suggest avenues for why we may see inequality arise even early in life. It’s obviously extremely hard to separate correlation from causality here — there are other differences across families — but this evidence is certainly suggestive that talking regularly to our kids is important. What this research doesn’t say is that you should narrate every diaper change. And it definitely doesn’t say that quieter parents are doing something wrong.
I am, in general, a huge fan of the use of data in parenting. There are situations in which good data are tremendously valuable. An example is early allergen introduction. Within the last decade, new research on the question of how to best reduce allergies has made it clear that introducing common allergens — peanuts, eggs, dairy — at very young ages dramatic reduces the risk of allergy development. Exposing children to peanut products at 4 to 6 months, rather than waiting until 12 months, lowers the risk of developing a peanut allergy by perhaps 70%.
This is an example where the effects are important, convincing, and large. But there are many places where the data are just less helpful. They’re suggestive, but not conclusive. Or the impact is minuscule. The size of the possible benefit to your child from narrating diaper changes is vanishingly small.
Despite this, so much data-driven parenting advice fails to differentiate between things that could make a big difference and things that should be determined by our preferences, our constraints, and whether we actually want to discuss poop with our infant. The result is that parents feel pressure to do things that could never have more than an extremely tiny benefit.
Sometimes, our desire to use data — to overuse it, really — is further confronted with the reality that the data can be simply wrong. Remember the study that suggested listening to Mozart helped students perform better on tests? How many people played classical music to their womb, or bought Baby Mozart videos? How many parents played Bach in the car when they would rather have had the Beatles?
Even if the findings had been replicated, this was an overreaction. And, in the end, the study didn’t hold up. It turns out that music might improve test results a bit — perhaps because it relaxes students before a test — but it doesn’t matter if it’s classical or not.
So, we sometimes over-interpret data. So what? We skip the Beatles, talk more than we want, occasionally find ourselves at a loss for words in response to a math test. But, really, those are small impacts.
Where I think they become larger is when we start to distrust ourselves, when data-driven advice generates anxiety. We all want to be good parents, and we don’t want to mess up our kids. “Following the data” seems to extend a reassuring hand. But as parents hear more do’s and don’t’s, there is more pressure, more ways to fail.
Parents don’t need more ways to feel like failures. Sometimes, we just need to hear “Good job.”
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Emily Oster is a professor of economics at Brown University. She is the author of 'Cribsheet' and 'Expecting Better.'