As a genre, parenting books generally don’t give their readers much room to think through what’s best for them and their children—they offer plenty in the way of “how to,” but little in the way of “whether to” or “why to.”
“By not explaining why,” writes Emily Oster in her new book, Cribsheet: A Data-Driven Guide to Better, More Relaxed Parenting, From Birth to Preschool, “we remove people’s ability to think about these choices for themselves, with their own preferences playing a role.” Oster is an economist at Brown University, and Cribsheet is her extensive analysis of what research has to say—and perhaps more importantly, what it doesn’t have to say—about the upsides and downsides of breastfeeding, potty training, and circumcision, among many other issues that come up in the first few years of a child’s life.
In Cribsheet and its predecessor—Expecting Better, Oster’s 2013 book about pregnancy—she describes a decision-making process for parents that’s informed by economic thinking. She makes passing mention of decision trees and “Bayesian priors,” but the process is simple: Gather the available data; estimate the costs and benefits of each potential course of action; factor in your family’s preferences and constraints; decide. Helpfully, Cribsheet takes care of the first and second steps of that process—which can be the hardest for parents—equipping readers with the information to make decisions confidently (and early, in advance of the chaotic, sleep-deprived months that a newborn brings).
Reading Cribsheet has a soothing effect, for it stresses that there is not a single optimal set of choices about child-rearing. I recently talked with Oster about the messages parents get suggesting otherwise, as well as the limits of what parenting researchers can actually measure. The conversation that follows has been edited for length and clarity.
Joe Pinsker: Searching online for guidance on parenting can often leave people more confused and stressed out than when they started. Why do you think the internet is such an exasperating place to seek parenting advice?
Emily Oster: I think we are increasingly used to looking to the internet for answers in many aspects of our lives. It’s where I would go if I wanted to know, Who was the guy in that movie that I can’t remember? It’s also the place where I go when I want to know, Should I sleep-train my kid? I think the problem in the case of parenting is that, particularly when people are looking to make choices based on evidence or data, it can be very difficult to get a complete picture of the facts based on what people are writing on the internet.
In part that’s because there are a lot of different pieces of evidence in most of these cases: You can ping-pong between a website that says You must make the following choice because that’s the only choice and one that says That’s the worst thing you could do for your baby. I think the variety of very strongly held views is confusing and can leave you sometimes worse off than you were before.
Pinsker: In the book, you also mention the tendency of advice givers to avoid cognitive dissonance—if parents made a certain decision about raising their child, they want to believe it was the right one, and so they prescribe it to others.
Oster: I think that happens on and off the internet. Part of what happens on the internet is that there are many more people making those kind of arguments. I think it’s exactly the last step of what you said that’s important: It’s right for me, so it must be the right choice, as opposed to a right choice. I think that’s where we get into these judgmental attitudes toward other people’s parenting if they don’t make the same choice: They must have made the wrong choice, because I must have made the right choice.
Pinsker: It seems like one by-product of these mixed messages is that parents, in many cases, end up with very little information about the actual consequences of any given decision. Do you think this lack of clarity in part explains why even small individual decisions can seem so important?
Oster: I think that part of what happens in parenting is that your kids are very important to you, as they should be. When you make choices, you want to make the right ones, and because of this rhetoric around a certain choice being “right,” every choice ends up seeming very consequential. I think all these choices really take on a salience and importance that they simply do not have.
Pinsker: It seems like it’d be easier to draw straightforward conclusions from research on parenting if there were causal relationships in the data, instead of just correlations. But a big hurdle is that most parents aren’t going to willingly manipulate their children’s lives for the sake of some research project. Is there a fundamental mismatch between people’s hunger for a correct answer and research’s ability to provide it?
Oster: Yes, this is a very challenging literature because, as you say, parents are not randomly manipulating their kids. They’re also not making their parenting choices randomly, so when we look to do research on this, the natural way to approach it would be to compare parents who make one choice with parents who make a different choice. But since the choices are not made at random, the kinds of parents who make them are typically very different—for instance, in terms of income or education.
The fact that the choices are different and the parents are different makes the literature ultimately unsatisfying on many things. Not everything, but it’s frustrating for people who want to be making evidence-based decisions all the time. Part of it is recognizing there are limits to the evidence and sometimes you have to make decisions in other ways, like simply doing what you think is best, without data.
Pinsker: I wonder whether the underlying problem is that there’s a disconnect between what parents’ end goals are and what research measures. Parents’ goals tend to be qualitative—they generally want to raise resourceful, well-adapted kids—and those are things that researchers would have a hard time quantifying.
Oster: Yes, it’s interesting, and I would take the problem even one step further. Even if your only goal is to optimize your kid on some measurable dimension and get them the highest score on whatever test researchers are using, the data isn’t really that good at helping you figure out which choices would do that, but you can take a further step back and say, Actually, my goal isn’t to have the kid with the highest IQ necessarily, but to have a kid who’s a happy, well-adjusted, productive person. So not only do we not know what does it—we don’t really know how to measure it.
Pinsker: The inspiration for Expecting Better was that you were about to become a parent, and now your kids have grown up a bit. Do you expect that you’ll write a book like this for each phase of their lives? Or is this the end of the line because, as you note in the book, as kids age, there’s more variance in what’s going on in their lives, which makes the data even harder to draw conclusions from?
Oster: Yeah, I think this is the end of the line, although I actually said that after the first book. I found even in the age range covered in this book that as I got into topics more about older kids, there is some good evidence, but that same quality of evidence is not there—so it would be hard to write the next one. Maybe I’ll write it about some other subject entirely.