Think Better with Emily Oster

Jane Risen:
We're particularly honored today to welcome Emily Oster for this conversation and Q&A. So let me give you a little bit of information about Emily. In addition to being a professor of economics at Brown, Emily is the founder and CEO of ParentData, which is a service that provides a data-driven guide through pregnancy and parenthood. She's given countless parents the knowledge to confidently make informed decisions, and her newsletter now has hundreds of thousands of subscribers and a very tight-knit internet community. Emily's first book, Expecting Better, challenges conventional pregnancy customs, taboos, and norms, giving parents the information they need to make the best decisions for their own pregnancies. Next came Cribsheet, her second book, which brings the same data-driven approach to tackle the early years of parenting, and then in the Family Firm: A Data-driven Guide to Better Decision-Making in the Early School Years, she focuses on parental relationships with school-age kids.
Emily has been featured on many shows, including The Daily Show, which was one of my favorite moments, and PBS NewsHour. I should also mention that Emily just happens to be one of my closest and dearest friends. So we met in college, she was a bridesmaid at my wedding, and I got pregnant only a few months after Emily did, and I started going to her with my questions instead of my doctor, and although I may have been the first to do this, I am not the only one anymore, lots and lots of people are going to Emily to get their answers, and we will have time for all of you to pose your questions so that you can get your answers from her too. We are just delighted to have her here. Please join me in welcoming Emily.

Emily Oster:
Thanks, Jane.

Jane Risen:
You're welcome. So this is fun. So I would like to start because I have more background knowledge than the audience, but I thought maybe we could start by just asking you about your journey in the sense of where you came, from being trained as an economist and then turning to write books for a much more popular audience. And I'm curious if you could share a little bit about what motivated you to start writing books, and how has your training shaped your approach to trying to write to a more general audience?

Emily Oster:
Sure. So first of all, thank you, guys, all for coming, and thank you to Jane and the decision, this is such a treat to get to do this, and it's especially a treat to get to do this with you. So thanks for hosting. Yeah. Okay. So I'm trained as an economist. I love data, I love economics. My parents were economists, I'm married to an economist, it's a whole situation. And we really have always tried to engage with decision making about life through that lens of thinking about data, and thinking about what are the risks and benefits and how do we trade off, that's how my brain works, that's how my family works. And when I got pregnant in 2010, I assumed that is how it would work too. I had a kind of expectation that I would take the same, I don't know, spreadsheet, data-oriented knowledge that I took to house buying and it would go into my pregnancy.
And I think the one big moment was this realization that that wasn't how it was working, that wasn't how the books were, that wasn't with existing books, and that wasn't really how my medical experience was. And I spent a lot of my pregnancy, and yours, thinking about what does the data say about this question about what kind of prenatal testing should we have, and what kind of birth do I want to have? And so that was the origin of what made me excited about this. And the other thing that's probably useful is I really like writing. I like doing economics, I like doing data stuff, but I also really like writing, and I really like the challenge of... or the interest in trying to write about research for an audience that is not steeped in that.
When we are academics, we spend a lot of time, in an audience of academics, with jargon and with an audience who really speaks our language, and the translational challenge of taking that and then saying, "How can I help people engage with this even though they don't have a PhD in this?" And so I like that challenge, and at some point in the middle of my pregnancy, and this part is a little bit fuzzy because of pregnancy brain, I thought, "Okay, I'm going to take the things I like to do, I'm doing all this research, maybe I could write about it, I'm going to write about it." And then it turned into this book proposal, and then later, there were some books, and now, we find ourselves here.

Jane Risen:
You set up one of my other questions really well, actually, two, so I'm going to have to decide. So one of your gifts is being able to translate these really complex ideas about data for an audience that isn't very sophisticated necessarily about research, and statistics, and things of that sort, what do you have in mind as some of the key ideas that you think everybody needs to keep in mind when either a new study comes out, or probably, more likely, it's like the headline comes out and there is some information behind it but you don't know what it is?

Emily Oster:
Yeah. I think there's basically two things, and when I spend a tremendous amount of time on Instagram talking people down from Panic Headlines, there's two pieces. So one is just correlation is not causation. An enormous share of the Panic Headline on the internet is like we have two groups of kids, one of them watched five hours of TV a day before they were one, and the other group never watched any TV, and they turned out different, and it must be the TV, and you're like, "Come on, guys, please." And so there's a sort of correlation is not causation that I repeat over, and over, and over again. I think the other piece that I often want people to recognize is that the headline about data is almost always, "New study shows," with the implication that, now, we're going to use this piece of... now, this is our leading piece of information, and that's almost never true.
On all of these questions, there's a bunch that came before it. And I often want people to stop before the headline, before you read about the new study, make sure you know about the old studies, and then you can incorporate this rather than just saying, "This is the one piece of information I need to know," because usually, that's not true and that's often not helpful. So those are the two. And then look at the appendix tables, but that's a more niche... it's like a more niche comment.

Jane Risen:
Yes. The other thing that you queued up in my mind from your first answer was that I can specifically remember during my first pregnancy deciding to eat deli meats because you told me it was going to be okay. And I don't think I would've felt comfortable doing that if not for you showing me the data that compared, "You can avoid the deli meats, but you're going to have to avoid the lettuce too," and then it became, "All right, deli meats it is." I was curious whether you yourself can point to specific different choices that you made over the years because of the data that you found, that you changed the approach that you would've taken.

Emily Oster:
Yeah. So I think there were many things like that in pregnancy, in part, because I was doing all of this research in the service of there are specific things even in this book where I did the research for me, but some of them I did for you, for reals. And so pregnancy was all about that. I think one that comes to mind, by the time I was writing Cribsheet, I was on my second kid, and so I had made many mistakes already with the first one that I was working on rectifying, but the one thing that came out there was I spent a lot of time on toddler discipline and thinking about what kind of, I don't know... anyway, I ended up reading all about 1-2-3 Magic and thinking about consequences versus da, da, da, da, and that was one place where I wrote the chapter and then I brought it to Jesse and I was like, "I wrote this and let's change how we're doing it, because I think actually this would be a better approach." So that was one.

Jane Risen:
Yeah. Yeah. One theme that runs through almost all your books, I would say, is that individuals and families don't necessarily have the same preferences, the same goals, the same constraints, the same tolerance for risk, whatever the case may be, and yet you still want people to be able to rely on research and data to make their decisions in an informed way. And so I'm curious how can people do that given all of the differences across? And then my follow-up is that I'm curious whether there's anything that you would say... any advice that you ever give universally?

Emily Oster:
So okay, I'll come back to the second one. I think there are two things that are really hard about incorporating preferences. To set the stage, I generally think the right way to make most of these hard parenting decisions is to think about your question, figure out what the information is, what's the data that I need, and then combine that with your preferences to make some choice. And depending on what your preferences are, that means the same data is not necessarily going to lead to the same choices, and that's fine. I think what's very hard is when you make a choice and then someone else is like, "Why did you make this stupid choice?" And it's hard at the playground, it's harder if it's like your mom, or your mother-in-law. And so there's this challenge of holding in your head like, "I made the choice that's best for me, and that is not the best choice for someone else." And it's hard to take it if... but it's also hard not to judge other people. And it's hard not to be bossy.
It's hard... because you're like you think even if it's coming from a good place, it's like you don't want to boss other people, but of course, you do want to boss other people. So this is something that I think about a lot, is how to be careful not to tell people what they should do. And I think it's gotten... from the position that I'm in now, a lot of people are just like, "Just tell me what you did. I trust you, I like your books, just tell me what you did." And I actually am quite careful about not being like, "I can tell you, but it's not important, because it's not actually necessarily what you should do." Is there anything that you should definitely do? Yes, it is introduce allergens to your child early. But that's like... there's nothing... that's like, okay, and then no one's like, "Oh, I don't want to do..." It's like, yes, that's something you should do, but that's not a very... there's nothing big, there's nothing big and contentious where, I would say, the data is very clear, or very obvious.

Jane Risen:
I'm reminded your answer is reminding me of one piece of data that you did do in this early thing, was looking at natural childbirth versus epidural-

Emily Oster:
[inaudible 00:11:00] bring this up, I didn't know if you wanted me to bring-

Jane Risen:
... and all of these things. And so we looked at the very same data and Emily decided to go the natural birth route. And I am the Jane in the book that says, "You're crazy. Because I looked at that very same information-

Emily Oster:
It's like the day after you had your kid, you called me and you were like, "You're crazy." You're like, "You're so stupid. The epidural was amazing." Okay.

Jane Risen:
I was a big fan, big fan of the epidural, so much so that I had a specific preference, I'm probably the first who's ever gone in and I was like, "No, no, no, I already know which one I want."

Emily Oster:
Beforehand, you were like, "What if it doesn't work? How should I prepare for it?" I was like, "Well, it almost always works." You were like, "But what if it doesn't? I really need it." I was like, "It's okay."

Jane Risen:
Different preferences, right? Pain, yes, you run all the time-

Emily Oster:
It's like an endurance sports.

Jane Risen:
Exactly.

Emily Oster:
You don't like endurance sports?

Jane Risen:
I don't like endurance sports.

Emily Oster:
It's not for you.

Jane Risen:
No, no. So your work sometimes challenges conventional wisdom, and I was curious if you could share some of the examples of where you've come to a different conclusion than what conventional wisdom may have been, and how you respond to critics who don't like moving away from conventional wisdom?

Emily Oster:
Yeah, I think there's a few, like in the context of pregnancy, the thing that would... alcohol and the occasional drinking during pregnancy was the thing that was most controversial when that book came out, when Cribsheet came out, I think the most controversial thing was this breastfeeding is not necessarily the most important thing that ever happened and is not going to lead to all... like many of the benefits of breastfeeding are overstated. And those were both things that in their own way came with a fair amount of pushback. And I think one of the things I learned quite a lot over time is how do you talk about complicated stuff like that with nuance with front... making it clear, I thought, why that is my read of the data, and what it is that I saw in the data that challenged the wisdom, and why I think that the existing conclusions are not correct in many of those cases, while at the same time acknowledging some of the things that have come before?
And so there's... I learned over time a little more about how to talk about that. I think when you talk about stuff like this, of course, there's a fair amount of pushback in various ways, and there are pieces of that pushback that I think are very important. I think it is really valuable and I've learned much more over time to hear out the doctors that don't agree with me or researchers that said, "I have a different read of this data. We can engage on how much omitted variable bias do you think there is in this study. And I have a perspective about reverse causality and you have... but we come together." I think the piece of it that's very frustrating for me that I've never really managed to get past is the credentialing piece, is the piece where people are like, "I don't think you should be able to engage in these questions because you are not a doctor," and I think that I don't like that.

Jane Risen:
Yeah. Or the wrong kind of doctor anyway?

Emily Oster:
Yeah, the wrong kind of... Yeah, I mean, let's not... I'm not a real doctor, but this feeling of I have a particular kind of expertise and we could come together on facts and talk about the substance on which we disagree. But it's very hard to hear like, "I just don't think you should be able to comment on this because you don't have these letters after your name."

Jane Risen:
Yeah, I don't know. All right. Well, let me move to a question where you should be... this is your wheelhouse, not theirs, this is yours. When you discuss research, you sometimes explain why the data isn't as good as it needs to be to come to the conclusion either that they came to or the conclusion you might want to come to, anything like that. And so I was curious if there are any topics that you feel are just that if you could direct research, that you would direct people to particular topics that you don't think have been explored well enough?

Emily Oster:
So I think there's... I mean, I always want more research on every tiny topic, I think there's two spaces. So one is I think in women's health, we have underresearched people's experiences. So when we do research... and this is an inherent feature of research, we tend to like things where we can measure the outcome. So is it readmission to the hospital? Is it this kind of... like something... medical stuff tends to be easy to measure, but we don't often have good ways to measure how happy is someone basically. And so I think those... I wish we had a better... it's not so much a topic, but I wish we had a better system. I would like to be able to say, "How are we collecting information about how people are doing in those first months after their kid is born, and what are the things we could do not to make them not show up at the hospital?"
Yeah, it would be great to not have people show up back at the hospital, but how are we figuring out how to support people so they are happy, and so they feel comfortable, and they feel confident, we don't have that measure of that stuff, and I would like more outcomes there. Being in the same vein, the other thing that would be great is to understand better the role of parenting in kids' success. That sounds really vague, but there's a policy question here, which is, by the time kids show up at school and kindergarten, we see enormous socioeconomic differences in how kids are doing. And so there's something happening between zero and five, and I think some of it is about parenting, and I don't think we really could say what it is, what are people doing differently? And that feels like a very rich research space.

Jane Risen:
My follow-up question was... maybe it's for one of these topics or maybe it's something else. If you could have the resources and buy-in to run just a study that tackled something in a way that we haven't been able to yet, maybe trying to get around the causation and correlation confusion that people have, or something like that, is there a study that you wish... even if it's like, maybe practically, it's impossible, but in theory, this would help us answer something that you feel like we currently either don't know much about it, or we are drawing our conclusions from the wrong type of data?

Emily Oster:
I'd like to do a randomized study on supplements because I think they're stupid. And so if I had an unlimited... I'm sorry this is a very... but I think if I had an unlimited budget, I would just randomize a ton of people into these ridiculous supplements and then I would just show that they didn't matter. But it needs big, because you need to be really powered to reject small numbers. So that's for me.

Jane Risen:
Yeah. Yeah. You primarily offer advice for individuals and families to make the informed decisions within the systems that they're already living in, but you already alluded to this idea that maybe the systems could change. Maybe there are things that we could do that would make parenting easier and better across the board. And so I was curious if you had particular policy, initiatives in mind, whether it's in health, or education, or something, where you think we could change the system so that people had an easier time navigating the system?

Emily Oster:
I think here, actually, it's really important to make a very crucial separation between almost where these conversations are happening and where the policies matter. So I have a lot of conversations with people about how can I do optimal things... how can I optimize around my kid, and how can I make exactly the right choices for their education, exactly the right choices? And in some ways, so much of what I spend my time telling people is once you have come to ask me this, you're already doing the core things that actually really matter, like your kid has enough to eat, they have a stable place to live, you're probably reading them some books, they're not having a lot of adverse childhood experiences. You're already doing it, you're already doing most of the thing.
And so you're going to come to me and ask, do I need to rotate the Montessori toys? And it's like, "No." And so I almost think there's this piece where we're thinking in that space about, "How do I do parenting the best," where it's like everything is fine, all those choices are basically fine. And then there are a lot of people in America, and broadly, but in the US, where we are actually struggling to help them meet some of these more basic needs. And there, I think we can ask, "What can policy do to help?" And so I think there are some things that are pretty straightforward, like paid family leave would be great, that's pretty popular in a bipartisan sense, it doesn't seem to have done anything but more paid family leave, some support for child care, for high quality child care, which is not like $5,000 for each person, because that's not what child care costs. It is like ongoing support and support for the supply side of child care.
And then I think probably some kind of parenting, coaching to help people figure out what are some of the pretty basic research things we know, so serve and response, some of the basic stuff that is really implementable but we haven't told people about. And I think that's... if we could do a little bit more in that direction, I think that that's where I'd push policy.

Jane Risen:
One thing that I'm not sure if everybody is as familiar with, because you've mostly been in the parenting space, but during COVID, you were very involved in collecting data, and that has led to a lot of data on schools and all sorts of things. And I wanted to know if there's any lessons from the school data that you think are worth sharing to an audience that may not know that work?

Emily Oster:
Are there any... other than COVID doesn't spread in schools. You should open schools, they're important. I've spent a lot of time now in the school test score data and trying to think about what we can learn from that, the biggest takeaway if you look at how kids are doing in tests is actually back to the last answer is how much heterogeneity there is across racial groups, across demographic groups, across locations, and how kids are doing, and how much there is left to accomplish. In the Detroit public schools, there's like 10% of Black kids are reading at grade level, or passing the test scores... 10% of kids are proficient. That is a fact that people should just know, that's not okay. I basically think what we get out of this test score data, a lot of it is just like we are failing a big chunk of America's kids. And I think the data is important partly because I think if we put data like that in people's faces, maybe it will make it more visible what we need to do.

Jane Risen:
So you started when you were pregnant with Penelope, and she's now a teenager.

Emily Oster:
Yes.

Jane Risen:
So now that you have a teenager-

Emily Oster:
[inaudible 00:22:43].

Jane Risen:
Yeah, it is disturbing. And they're so tall.

Emily Oster:
They're so tall. I know. I know. She puts on these heels and she's like, "I'm taller than you." I'm like, "You're not taller than me, not yet. Not yet." Is yours as taller?

Jane Risen:
I am well beat. I am definitely... I'm gone.

Emily Oster:
You're behind.

Jane Risen:
So I was wondering if there are new topics that you are digging into because of the place that you're in, and I wonder if there's a book that's going to follow so that all of us know how to raise our teenagers?

Emily Oster:
I think that raising a teenager is very hard because it is not very amenable to data. It's not. When you're dealing with a little kid, there's a lot of these, I don't know, the questions are more... they're less heterogeneous. When I wrote Cribsheet, it was not hard to outline the chapters, like the you have a baby, they need to sleep, they need to eat, they're pooping, they got to eventually poop in the potty, basically, it's pretty clear what's going to happen, and where you need to go from A to B. And when you get to an older kid, the problems they have are more different, and your techniques for dealing with them are... you're constantly trying to reinvent different strategies and figuring out how to parent someone who looks like an adult, they're taller than you, but also does not have a fully formed frontal lobe. I have one amazing parent... Would you like to hear my best parenting strategy?

Jane Risen:
Yes.

Emily Oster:
Okay. It's this-

Jane Risen:
Of course.

Emily Oster:
It's this. I sit in... my kid's rooms, there's a den and then there are bedrooms. And in the evening, after dinner, when they're getting ready for bed, I just sit in the den on my computer, and then if my daughter happens to come in and say she wants to talk to me, I immediately close my computer and pretend that I wasn't actually doing something, and I'm just there to talk to her. And if I'm just sitting there, sometimes, she will show up. That's it. That's what I have. And it's really time-consuming because I'm there for two hours, it's fine, I'm working, it's okay. But that's it, it's just like being there. There's so much more time that you need relative to when they were little, which I find surprising. So therefore, I'll not write a book because it would just be that one page.

Jane Risen:
That's it?

Emily Oster:
That's my only tip. I got that from someone else too. They already have a book about it. So I'm done.

Jane Risen:
We have a few more minutes before we're going to let the audience here join in with some questions. But we also had some of our Zoom questions come in, and so I thought I would be able to ask a couple of those for our audience at home. And there were a few themes that came through. One, I liked a lot, which is basically just to remember that we're in lots of roles other than parenting roles. And so one of the questions was specifically, "Any advice for grandparents, mothers-in-law in particular?" Which I enjoyed.

Emily Oster:
Yeah. Oh, my god, so much-

Jane Risen:
And this isn't necessarily just for your mother-in-law.

Emily Oster:
Right. My mother-in-law is wonderful and wouldn't listen to me if I said, which is perhaps the point. I would say, the best... I mostly hear from the children-in-law about what they wish, and I think that the main thing they wish is to not hear advice, and so maybe zip it. But I actually think one thing that I would say is people are often willing to hear a piece of advice once, but don't say it multiple times. You've said your thing, you said what you think, don't say it again.

Jane Risen:
I would like to also, I think-

Emily Oster:
Do you have some advice?

Jane Risen:
No, I was going to say, for seminars, I think that's another good... you asked your question-

Emily Oster:
Don't ask it again.

Jane Risen:
That was it, you got your chance.

Emily Oster:
You got your chance, you just get one. Just get one.

Jane Risen:
Another person wrote in saying, "Parenting one-on-one strategies don't work for me. My kids are neurodivergent. How can I keep my cool admits standard parenting advice?" And I think you help people a lot with this.

Emily Oster:
I think that is... I mean, that's a very hard space, and I think some of it is about finding... and I think this is true, no matter what is the particular thing with your kid, is finding a source, or a method, or something, whether it is a book, or the provider they're using, find something that is going to work for you and then try to keep that. It is very easy on the internet to be influenced by the latest... "If only you did this thing, you would immediately fix your problem," and that is definitely not true almost always. And so I think just sticking, figuring out something that is going to work for your family, and then trying not to be influenced by the other random stuff that people are telling you to do on the internet is about the best you can do.

Jane Risen:
A lot of people wanted to know about screen time.

Emily Oster:
Yeah. Did they? Did they?

Jane Risen:
Yeah, they did. How are you going to solve this?

Emily Oster:
Solve it. There's two different screen time issues. I think there's a narrow, fairly simple-to-solve screen time issue, and then a harder one. So I think on the narrow issue is like what do I do about screen time with passive screen time with my small children? How should I navigate my three-year-old's consumption of Ms. Rachel. Ms. Rachel... I don't know, my kids are too old for... anyway, so how should I navigate that? And I think I have a very simple way for people to think about this, which is just when your kids are watching screens, they're not doing something else, and there's various other things that are valuable and you want to have in your day, probably having a meal, and sleeping, and being outside, and when they get a little older, doing their homework, and thinking about those are things you want. And then screen time is also part of your day, but you should plan when the screen time is, and it should not interfere with the things that are more important.
So if your kid is watching screens instead of sleeping, that's bad because they need to sleep. And so if you think of screen time as not a good or not a bad, as part of your... thinking about in advance, this is when the screen time is going to appear in our family's life, and this is the limits we're going to put on it, and then holding those limits. It's not that it's easy to do those things, but that is a good framework that can work and can generate a relationship with screen time that you are comfortable with. Then there's phones. And then your kid is older and they want access to phones. And Jon Haidt said that if you give your kid access to phones, they're going to have a mental health problem. And this is basically where a lot of people on older kids are.
I think that's this hard because I really think Jon is great, I think he has overstated the quality of the evidence about the relationship between phones and mental health, it's come out to seem like the entire rise in mental health problems for kids are exclusively due to Instagram, and I think that's an overstatement of how important Instagram is. But also, it's pretty clear that some aspects of these things are not good. And so it is a very hard thing to have advice on other than to say you are the parent and you have to be willing to set boundaries, but also somehow teach your kid to eventually set their own boundaries. I don't know.

Jane Risen:
Nope, that's good.

Emily Oster:
Which is not very easy.

Jane Risen:
Maybe I'll give the final Zoom question and so then we can turn to the audience after this one, was someone wrote in saying, "I'm going from one kid... or one to two kiddos in July. Best tips?"

Emily Oster:
Prepare your partner. I actually think that if there is one more defaulty parent, when you have... or even if there's not, when you have one kid, one person can be off at any given time. And when you have two kids particularly at the beginning, no one can ever be off. And I think that is a difficult adjustment. So I think there's nothing particular... you already have this stuff, the second kid's usually easier than the first because you kind of know the situation. But I would say prepare your marriage, don't worry about your house. Amazon Prime can deliver stuff if you forget.

Jane Risen:
Very good. I think we're now ready to open it up to the audience, so if you are interested, raise your hand and we have people... That was very fast, I love that. We have mics coming so that you can ask your question.

Speaker 4:
Thank you. I've given a lot of parenting advice myself in a 30 plus year as a pediatrician, and I agree with you that much of what we tell parents is style and not really important and lots of choices. And you were asked at the beginning, what's the one thing that you would share? And you said allergen introduction. And there's another thing I would like you to say-

Jane Risen:
Which is?

Speaker 4:
People are dying of measles in this country.

Emily Oster:
Oh, yeah. So I would say, in the book, in Cribsheet, which I wrote before, we had so much measles, I talk about the two big, "Here is what to do chapters," are vaccines and trying to talk people down... I spent a lot of time obviously trying to talk to people about what is going on with the measles vaccine, how they should think about measles vaccine safety. So yes.

Speaker 4:
And not just the measles vaccines, vaccines in general.

Emily Oster:
Yes. I spend a lot of time talking about measles... about vaccines in general. And I think-

Speaker 4:
And you have a platform, and you-

Emily Oster:
And I spend so much of my time on that platform talking about how people should vaccinate, and misinformation about vaccination. And I think there's a lot of space to convince people if we hear out their concerns and try to meet them where they are. So yes, I will say I really do spend quite a lot of time on my platform talking about measles and other vaccines. I promise.

Speaker 4:
Thank you.

Speaker 5:
Hi. How do you-

Emily Oster:
Hello. Sorry. Where is this?

Speaker 5:
How do you teach a teenager or a child to find out what is correct, what is true today? Some information can be found out in an Emily Oster book, some information can be found out on Wikipedia, some by asking ChatGPT, for instance. But these days, there's so... exactly, there's so much just sources of information. How do you teach somebody, a kid, which one of these sources is the correct source, or something like that?

Emily Oster:
I think it's a hard... it's interesting, no one has ever asked me that, which is good. I think part of what's hard about this, I think generally, it's not that you could point to the encyclopedia of the sky and be like, "This is always where you can look." I think what we can teach kids is, "Here are some trusted sources and here is a general sense of where to look for information." And then I think we can teach them the core thing, which is, "You should probably never have just one source of information, and here is how you engage with generating knowledge and almost checking what your sources are." Because I think ultimately, the way that we get to understand the truth on a lot of things is to see what the other side says. And so there's some piece in there. But I don't know, that's a very hard question. Hard and good.

Speaker 6:
So my question is related to balancing autonomy and safety, especially when your partner and yourself have perhaps different ideas of the ideal balance.

Emily Oster:
Okay, that's hard. I think there's a generic piece of advice here, which you probably know but I will repeat, which is having those conversations in a moment in which you are ready to have the conflict and try to get to a solution is going to be more effective than having the conversation with your partner when you are disagreeing about what the person is doing right this very second. And so I think as much as possible, confronting that disagreement and saying, "Okay, we are trying to do this activity of raising this person together and let's actually try to figure out can we come to a compromise on this?" Because if you don't, then you will just have this conflict over and over again. Then there's a question of what is the evidence about balancing safety and autonomy? And I think this is something almost everybody struggles, a lot of parents struggle a lot with, in part, because it's very easy to see the safety risks, and it's very hard to see the risks of not letting your kids do things, particularly in the physical world.
It's easy to see why I would be afraid to let my kid walk home from school by themselves, and it seems like I'm trading that risk against a not risk. And the phrasing of no option is completely safe can sometimes be helpful to think about the option of never allowing your kid to do this also has some risks of... they're a little more vague, and existential, and developmental as opposed to literally they're going to fall down and scrape themselves. But recognizing the risks of the other option is sometimes a helpful frame. And I will also say, I don't know which side of the autonomy safety you are on, but for many people, taking tiny steps is helpful.
So if you think, "Let's let our 6-year-old walk home from school by themselves," and your partner is like, "Are you kidding? We're going to be walking them home until they're 18." Maybe the first step is like, "Why don't we meet them half a block closer to school?" Or, "Why don't we let them go to the end of the driveway," and just if there's a little bit... because the first time you do something like that, it's very scary, and the third time you do it, it's not scary, and that adaptation can slowly build into maybe something that you're both comfortable with.

Speaker 7:
Hi, I was wondering if you could talk about a time when you read the data one way, and then later, changed your mind, and how you handled that, and what about it changed your mind?

Emily Oster:
You mean if I read the data and then new data came out, or I ultimately decided that the data wasn't the way I had read it in the first place?

Speaker 7:
The second one.

Emily Oster:
There are times in my academic work when I've made mistakes, and one very public one, very early in my career, which was closest to that, but I don't think that that has... and that was a very embarrassing thing, where I should have been more careful at the time. I don't think that that has happened in the parenting data space. New data has come out... there certainly have been cases like whether you can sleep on your back during pregnancy where the data has kind of flip-flopped some over time, but I thought that the data said this and then when I read it again... and I said that publicly, and then when I read it again, I thought it was different, I do not think that that has happened, which I... and now, probably once I leave, I'm going to be like, "Oh, I should have said this." I'm going to keep thinking.

Speaker 8:
Hi. So you said you have a lot of data from COVID times. So there's the disease itself, but then there was also the mental health impact of the social isolation, particularly for teens, pre-teens, in that cohort. So I had a teen who was ready to graduate high school, and I also had a pre-teen, and their development was significantly different, the pre-teen was more impacted mental health wise. So do you see any obvious data about mental health?

Emily Oster:
I don't think we have seen what we will eventually see about this. You can look at mental health trends over time, we definitely saw some blips up in the pandemic. It's basically returned to the upward trend that we saw pre-pandemic. So I think there are a bunch of questions like this which remain unanswered, and my guess is well, eventually, we will answer a bit more of them over time. But the kind of blip up during the pandemic doesn't seem to have persisted, it basically went back down, and then there's been a trend in deteriorating mental health for quite a long time.

Speaker 8:
Thank you.

Jane Risen:
I think here, we have one.

Speaker 9:
For older kids, how do you separate nature from nurture when you collect data and analyze data, besides just collecting huge amount of data?

Emily Oster:
I think even collecting a huge amount of data is basically impossible. Many of the questions that we would have about older kids about what determines success, like how do I determine... we will have no idea, because there's so many different factors and there are so many different inputs, and the things that happen are so rich and varied, and kids are so different that... this is part of why Jane asked, "Are you going to write another book?" And I think part of the reason older kids are really hard is the data is just really, really poor. It's already not that great when we talk about two-year-olds, by the time you're talking about 17-year-olds, we're kind of done. Yeah, sorry.

Speaker 10:
Thank you for coming over here. So I had a question on the involvement of grandparents. So if you look at the word parenting, it was never in the dictionary until the '70s. And I didn't believe it, but when I went and looked at a 1968 copy, it turned out it was right. So there was no word of parenting. The theory is that the involvement of grandparents has come down. So people have started looking for parenting strategies, because back in the day, there was always help required. So my question is what do you see in the research about involvements of grandparents, and how has the research outcomes changed, and do you see any good or bad trend because of that?

Emily Oster:
Yeah, it's interesting. I think we have seen a decline in multifamily households over time. And so that kind of grandparenting and involvement, which, of course, in a historical evolutionary space-time was obviously very, very important in some sense. The importance of grandparents is why women survive past reproductive age basically, it has to be. And so over the more recent past, people have... nuclear families have become more important people, there's more mobility, people have moved away from grandparents. And so I think that is a trend, I am not sure that's why this sort of parenting as it is currently practice has become more important, but certainly, this move away from the nuclear family is something that has... or move away from grandparents, is something that has happened over time.
The other thing that's happened I think is, to emphasize this shift, is that people are having kids older, and so their parents are older. And if you think basically grandma's job is going to be taking care of... my mom's going to take care of the baby, if I'm 40... if she's 40 when she had me, and I'm 40 when I have my kid, she's 80, and that's not going to work, or at least depending on the person, that's going to be tough. And so I think that's another piece of this, it's become more difficult to have that be your core childcare.

Speaker 11:
Hi. I just wondered if you could... if there are any spaces where you think that maybe you might not rely on evidence-based decision making, but I don't know, something like intuition, or value-based decisions might happen? And I'm not maybe thinking so much of parenting, but just other areas that you've looked at.

Emily Oster:
I think every decision has to involve your values and preferences. This is the... people will sometimes say this as if it is in conflict with the idea of using evidence, and I think that that's not the way I would think about it. So if someone says, "I am just doing this because of my gut," the way I read that as an economist is my preference for this is very strong. And so in some sense, you could still say, "I'm going to look at the data, and I'm going to think about what the data says, but then I'm going to combine it with my preferences, and my preference for this is really, really strong." And even though I see that having an epidural is terrible, I'm still going to do it because I hate... that's not a good example.
But I think there are cases in which people would say, "Look, I looked at this but I decided I really want to do this other thing." And I think that's a totally reasonable set of choices, and in fact, exactly how we should be making decisions. I always think it is valuable to have the evidence, even if you were going to say my preference, my preference dominates. And so I don't actually think that decisions based on intuition are some kind of alternative to evidence-based decisions.

Speaker 12:
So thank you so much, Emily, for being here today. This has been really helpful. So I remember reading during the midst of the pandemic that one of the unfortunate things was the educational development loss in children due to shutting down the schools, or once they finally opened, it's more virtual versus in place. So I'd love to hear your opinions on... or what policies have you really pushed for in order to recoup some of those losses during the COVID pandemic times? And also, just with the advent of AI and all of these really great AI educational-focused solutions, do you think that could be played or those could play a role in helping to recoup some of those losses? Thank you.

Emily Oster:
Yeah. So when we look at the COVID pandemic losses, the virtual schooling was very bad for the initial losses, and then subsequently, there's been a fair amount of variation in how much recovery there has been. And so there's a lot of variation across states in how much recovery they've seen, which actually is largely unrelated to their pandemic schooling mode. So the amount that they've recovered just doesn't... you lost more if you didn't open your schools, but the recovery doesn't seem to relate too much to that loss. There are some places that have been much more successful than others and I'm not sure we have a great sense of why, but to give you an example, there's a huge amount of variation in ELA, in English language arts. So we have English language arts and math. Every place in math is crawling itself back at a slow rate.
And in English language arts, we have states like Massachusetts, which actually is substantially worse off than they were even right after the pandemic. And states like South Carolina, which is substantially better off, or Mississippi, substantially better off than they were before the pandemic. So we're seeing big variations in recovery. I don't think we have a great sense of what's going on there. One thing that shows up in a casual read of the data, which I think in some ways relates to your second point, is some places have been teaching reading with phonics and some places have not. And the places that have been teaching reading with phonics for a longer period of time have recovered more quickly.
For me, that actually speaks a little bit to some of these AI questions, because I actually think the problems that we're having, like kids not being able to read, are not really going to be solved by AI, they're just going to be solved by coming up with a better reading curriculum and having enough people to teach the kids to read. And then from there, they can be equipped to engage with AI. But I almost think we're having problems that are before the AI problems, and it would be great to use AI for ed tech, and I think there's a lot of interesting promise there, but if you can't read, you can't engage with the AI because you can't write in questions. And so let's get the reading first and then we can talk about GPT.

Speaker 13:
Hello. It's exciting to ask you a question face-to-face on Q&A Wednesday.

Emily Oster:
It's just like, I know, it's just like Instagram.

Speaker 13:
This is a very personal question. I'm due in August.

Emily Oster:
Congratulations.

Speaker 13:
Thank you. Vaccine-hesitant family who will be visiting from across the country to help out, one of which is an educator. And so I am wondering your advice on where the hard line may be, even though it could be personal preference, for vaccines to really push for people to get in the first six months or so?

Emily Oster:
Most of the vaccines... So there's two kinds of the vaccine hesitancy that I would think about here. One is I've got a bunch of... not I, but if you have a bunch of nieces and nephews who don't have any of their routine childhood vaccines, and I think that is a place I draw a pretty hard line for the first several months at least, because you really don't want a baby with measles, and you don't want a baby with pertussis. If what we're talking about is we have older family members who don't want to get the COVID vaccine, who don't want to get a Tdap booster, who don't want to get a flu vaccine, I think the sterilizing immunity of those is less strong, and so these arguments against having them are less strong. If I had to push for one of those, I'd push for flu, basically.

Speaker 14:
So when I was a kid, I learned cursive.

Emily Oster:
Oh, yeah, me too. What a disaster, right? You use that at all?

Speaker 14:
I use it all the time.

Emily Oster:
Do you?

Speaker 14:
Yes, I usually have notes, I usually take notes and everything. And I got to walk to school, so I mapped. And things that I have... I have an 18-month-old, things I've read about is, one, is the creative nature of cursive, and how you're using a different part of your brain, as opposed to either printing or also typing. And then to the other point of mapping your environment where either we're not letting kids go out far enough to really have that spatial awareness, or even something like Google Maps, which is mapping things for us. And so is there anything on that kind of... I don't exactly know the question, but the impacts of things like writing cursive, or mapping your environment, those kind of behavioral... like I said, I'm not exactly sure of the question because I don't understand the depth of it, but-

Emily Oster:
Yeah. So there are pieces of this. So the mapping environment is, for sure, true, and we know from studies on London cabbies that basically show you can see the way that their brains have map things is different because you have to memorize all of the cabbie routes, or whatever, in London. And so I think it's, for sure, true that people over time have developed less of that set of neuronal pathways that would be appropriate for mapping, not so much because they don't walk to school, but just because everyone is using Google Maps all the time and they can't get back to their house from three blocks away without Google Maps. And so you could say, "Is that a problem?" It's only a problem if you think there's something particularly special about that mapping, which I'm not sure that we know enough about the brain to know that that's true.
And I would say the same thing for cursive, I have never seen any data on the relationship between cursive and creativity. It's, for sure, true that if you look at the brain, if you said, "I'm going to teach somebody cursive for six weeks and have them really focus on it, I'm going to run them through an fMRI before and after," you're going to see things that change in their brain. But whether those are important, I think that's like we just don't know. There's a lot of plasticity in your brain.

Speaker 14:
We might find something else.

Emily Oster:
Maybe typing is going to be good for some other thing. I don't know.

Jane Risen:
I think we have time for one more question. Sorry to make you choose.

Speaker 15:
First. I want to say thank you. I was pregnant for the first time in COVID and you were so helpful for trying to parse through all the noise. So I have a two-part question. My first is, when you were starting your research for all of your books, is there something where you're like, "Surely, the deli meat thing is true," and you found out that was... the evidence convinced you otherwise? And then is there some conventional wisdom now that you really feel like you want to dig into and challenge?

Emily Oster:
So I was... what was the most surprising? I was surprised... the vague sense of where Listeria comes from, I found surprising. I thought that Listeria would be more concentrated. And then when I started looking at where Listeria is, it was just so many different random places. So that was one moment of early on where I was like, "Okay, I did not expect that." Although there were many surprises during pregnancy. Is there conventional wisdom I would like to think about now? I don't know, my brain is so often in older kid land these days. I'm not sure I have any conventional wisdom that I want to debunk right now.

Jane Risen:
All right, I am going to have to pause things here, even though I know there are many more of you in the room with questions. But for the sake of time, we're going to have to end here. We will be back next year in the fall, and the winter, and the spring with our Think Better series with a wonderful set of speakers. But for now, I just want to, again, thank the people in the room who came out in person, thank everyone on Zoom who joined, and most importantly, one last chance for us to thank Emily for coming and being with us.