How to Avoid the Worst Parenting Mistake

Emily Oster caters to a data-obsessed crowd of modern parents. But sometimes you just can’t optimize your kid.

Emily Oster is featured on a colorful background with graphics surrounding her.
Rachel Hulin; The Atlantic

Emily Oster is a popular target for irrational hatred. When I was reporting a story on how progressive communities have approached COVID-19 lockdown restrictions this spring, she showed me an email she got from a random person who had written to all of her bosses at Brown University, accusing her of promoting genocide. To be clear, Oster does no such thing: She’s an economist who has become semi-famous for her books on data and parenting decisions. Recently, she has also been a vocal advocate for opening schools during the pandemic, based on data she collected suggesting that in-person instruction can be safe with some mitigations in place.

Oster’s various projects are all arguably connected by a single worldview: Data are useful for guiding our choices, including what we feed our kids or whether we send them to school during a global virus outbreak. Her new book, The Family Firm, is an ode to rigorous decision making: Oster and her husband—also an economist—run their household like a business firm, she writes, employing many of the same principles she taught to M.B.A. students in her microeconomics classes.

Getting mad about the way other people use data to make individual parenting decisions is a little silly. But when the stakes are collective—when one mom’s choice to send her daughter to school could affect the health of another child—lots of people do get angry. Oster has come to represent a certain mode of thinking about statistics and risk that can feel alienating or callous to some people who are worried about COVID-19. But like a lot of us, Oster feels humbled by the pandemic. “I learned that while good logistics and decision-making helped us navigate this strange and hard moment, they couldn’t give me control,” she writes. “At every stage, I am surprised at how parenting forces me to recognize the things I simply cannot command.”

I spoke with Oster, who’s an occasional Atlantic contributor, about her new book and what data are really for. Our conversation has been condensed and lightly edited for clarity.


Emma Green: You start your book with the example of “redshirting”—holding kids back for an additional year before they start kindergarten. What struck me about this example is the loose, ambient objective associated with it: “What will help my kid do better?” What does it actually mean for a kid to “do better,” whether you’re talking about enrolling in kindergarten or something else?

Emily Oster: One way you could “do better” is on standardized tests—you could perform better academically. I think that is how people often mean “do better.” In this particular case, when people come to me with this question, what they mean by “do better” is mostly social or behavioral. What people are worried about is: Is my kid ready for the behavioral demands of kindergarten? Would they do better to wait until a time when they can sit still more?

Green: It’s interesting that you brought up test scores, because that’s a theme throughout your book. On a lot of the parenting questions you deal with, the available studies are centered on test scores. I assume that’s because they’re an easy form of measurement: You can collect a data set of scores and organize them neatly into charts and graphs.

I wonder if you think test scores are a useful measurement for making decisions about how a kid should have a good life.

Oster: You’re exactly right about why we focus on test scores, which is that they are something we can measure. Other things we look at are also things you can measure, like graduation from high school or enrollment in college. And I think it’s also true to say that is by no means the only thing we should care about. We should also care about whether kids are nice and happy and have relationships with other people.

Are test scores a useful measure? They don’t capture nothing. When we talk about measuring things like the quality of the school or how kids are doing or how a state is doing, I do think there’s value in test scores. The tests are not perfect. But it’s also not right to say that we learn nothing from comparing test scores across schools.

Green: I want to follow up on that, because measurable outcomes loom in the background of your book: test scores, future earnings potential, the age when a kid enters the workforce. It comes off as a form of class anxiety. People are seeking out data about particular metrics, which are mostly about keeping kids at the top of the meritocratic heap. Is that a fair assessment?

Oster: I guess I don’t think that’s unreasonable. When we think about “What does it mean for our kids to be successful?” often people think of that as “They’re going to achieve something like what I achieved or more than I achieved or have a comfortable life in some way.” That’s absolutely wrapped up in a lot of these class issues.

It is true that there’s a set of parents who are looking for data, and the thing that they’re looking for is “What is it going to take to make my kid successful in these measurable ways?” But I think it’s not right to say that those people don’t care about anything else, or that I don’t care about anything else. When we make decisions, data is one piece—it’s not the only piece.

Green: In your book, you veer back and forth between these two impulses you describe: On the one hand, you serve a specific kind of reader who values mastering the available data and academic literature on parenting questions. But you also gently push back on this mindset, trying to get people to see that data isn’t everything. Do you see yourself as the Pied Piper of data-addicted meritocrats, or do you see yourself as a guru coaching them up the mountain of wisdom? Something in between?

Oster: Oh my God, I don’t see myself as either of those things. Perhaps I should—they both sound amazing.

I do think that, for at least some of us, it can be very tempting to get wrapped up in some measures of success that are test-score-based and objective. That is often incomplete and, in some cases, counterproductive. I’m using this platform to say, “Hey, yes, you can use data to look at test scores and objective outcomes. But that should not be the only thing in the decision set. And sometimes that focus is totally misplaced.”

Green: Why do you think “some of us” get wrapped up in these objective measures? You describe yourself as one of these people: You once tried to research whether there was academic literature on whether your daughter should wear baby mittens. You were self-aware enough to be like, That’s insane.

Oster: Not in the moment! Later! In the moment, I was like, I need to download this paper.

Green: Well, lots of people do crazy stuff when they’re sleep-deprived with a newborn. Your sins are forgiven. Anyway, what is that draw to data, specifically when we’re thinking about these intimate questions about how we raise our children, how we structure our families, and how we relate to our communities?

Oster: I think it’s that we really, really don’t want to mess up. We want an answer for how to do it right. Somehow we got this idea that data can help us answer some of these personal questions and get them right. That carries over into parenting: This is a thing that feels like it should have an answer. It feels like data would give us control in an environment in which it’s easy to feel like I don’t have any control.

It’s a hard realization for this era of parenting that the data is basically never going to answer those questions for you. You’re never going to be sure that you made the right decision, especially in these bigger questions with older kids.

Green: The tagline of your book is “A data-driven guide to better decision-making in the early school years.” But it sounds like your spiritual takeaway from writing this book is “Data has no answers for us. Ultimately we must navigate the wilderness of parenting alone.”

Oster: It’s not totally that. I think it’s closer to “Data doesn’t have all the answers, and you need to be precise enough about your question to know where to plug it in.” Say my kid is struggling to learn to read. That’s a problem you might face. The question is “What should I do to fix that? Is he being taught to read in the right way?”

That’s something we have data on. When you approach the problem more deliberately, you will see where the data comes in. The deliberate piece just has to come first.

Green: You’ve gotten questions from parents about which extracurriculars they should enroll their elementary schoolers in to improve their chances of college admissions. You wrote that there’s something that feels faintly wrong about that.

What do you think drives people to think about college when they’re planning activities for their first grader?

Oster: There’s actually a pretty small share of people who have that in their mind when their kid is little. But I think part of what drives that is, if you’ve already had your kid enrolled in tutoring so they can take a test to get into pre-K, it’s easy to go all the way down the road of “We’re going to keep doing this, so how can I make it go as well as possible at all of the next steps that we have?” In some places in the U.S.—New York being the most obvious one—we’ve moved down to doing this when they’re 4. And then it’s easy to see why you’d assume that you need to make choices that would be helpful later.

Green: Why do you think your readers look to experts for guidance on some of the most intimate decisions about their domestic and family lives? In another time or universe, people might look to their pastor or their family members or their neighbors to help them with decision making about their family life. Why do people look to an economist they don’t know for that?

Oster: I think there’s a broader reason, having nothing to do with economists. We are, in a lot of cases, raising our kids in environments that are not the town we grew up in and not around our parents—environments that are more individualistic. So there is a search for sources of information or help with some of these decisions.

There’s also been a societal move toward the idea of data as an important thing that we know how to analyze and use to make decisions. Even in the decade since I published my first book, the idea of personal data has become a much bigger thing. I think that’s part of the appeal of the stuff that I write.

Green: Your scope of expertise has really increased during COVID. You’ve taken on a role in collecting data around the pandemic and school safety. You got lots of backlash, which partly focused on expertise drift—people saying, “You’re not an epidemiologist; you don’t work in public education; what makes you qualified to talk about this?” How do you answer that question for yourself?

Oster: We were talking about data. The question was “What does the data say?” And my training is in analyzing data. I also spearheaded the most comprehensive data set about this question. You could say, “I don’t think your data is perfect. I would like better data.” But part of what has been frustrating is feeling like people didn’t want me to weigh in on this because of the name of my degree. That is going to exclude a lot of people from a discussion where different perspectives are useful.

Green: Do you think that having ventured out of the academy to write a parenting book freed you up mentally to be willing to step into a discussion that would typically be considered outside your zone of expertise, at least in the academic world, where lanes are aggressively enforced?

Oster: Definitely. There are a lot of confines in being an academic, and a lot of judgment if you do things besides writing papers for academic journals. I had already done all that. I already ruined that for myself.

Green: Are you going to keep writing about COVID and schools?

Oster: Sort of. We’re working to put together a very big, comprehensive database on what happened with schools over the last year, to make it accessible to researchers. I doubt I will be as public with the discussions of COVID in schools in the next year. I don’t know if I have it in me to keep up with the battles.

Green: Is it a relief to be back in the parenting-advice world?

Oster: Yeah. I feel like I’m directly helping people. The reason I like writing the books is because I basically take an academic literature and distill it for people so it is more understandable. That’s something I really enjoy and that I get to do much more in the parenting space.

Green: You say you’ve sworn off writing more parenting books, but the logic of the publishing industry is that successful books must yield spin-offs. The best title I’ve come up with for your next project is Avoiding Bunk in Your Pursuit of the Bunk: A Guide to the Teenage and College-Admission Years. That’s pretty good, right?

Oster: It’s really good.

Green: You can have the title! But it will be on the record in this article that you stole it from me.