Sarah Gonzalez: I'm Sarah Gonzalez, in for Tanzina Vega, and you're listening to The Takeaway. President Biden has proposed the American Families Plan, and as part of the plan, he is proposing $200 billion for free, universal preschool. If passed, experts say the package, which would also set aside $225 billion to make child care more affordable, would represent the largest-ever American investment in child care and early education, and to help us understand the long-term effects of preschool, we can look to Boston.

Back in the late '90s, Boston offered preschool to some of its children through a lottery, and this week, researchers released a study looking at the long-term effects of preschool education on those children. Here to help us understand what they found is one of the authors of the study. I'm joined by Christopher Walters, Associate Professor of Economics at UC Berkeley and affiliated with the MIT School Effectiveness and Inequality Initiative. Christopher, thank you so much for being here.

Christopher Walters: Hi, Sarah. Thanks so much for having me.

Sarah Gonzalez: Let's start first with just what happened in Boston in the late 1990s with preschool, who was involved in this? What happened?

Christopher Walters: Sure. Like a lot of states and localities in the US, Boston has a public preschool program for three- and four-year-olds, and its program actually goes back quite a long time, like you said, back to the late 1990s, and as you alluded to, kind of interesting institutional feature of the way public preschool in Boston works is that assignment to the program actually happens partially by lottery. Parents who want their kids to attend preschool will submit their preferences over which programs they want to attend.

They have some priorities based on where they live, whether they have a sibling enrolled, and then within those groups, because there aren't enough seats for all the kids that want to attend, the District actually breaks ties by random lottery to decide who gets a slot. We as researchers, we can use that to understand the impacts of the program by comparing the kids who randomly win and lose those lotteries to get seats.

Sarah Gonzalez: Researchers like yourself thought, "Well, this is a good opportunity to follow these children and do a really long study." We'll get into some of your findings in a bit, but first, can you just give us a sense, what did past studies on the benefits of preschool show us?

Christopher Walters: We actually have a lot of pretty encouraging evidence on the impacts of preschool from a couple of different kinds of studies. One set of studies are randomized studies, kind of like what we're doing in Boston, in the past, typically conducted on pretty small scales. A famous example of that is the Perry Preschool Project from the 1960s, which randomized a very small number of kids to an intensive high-quality preschool program. Researchers there found very large positive impacts on all kinds of outcomes over kids' entire lives from attending preschool.

Then, we have studies of some bigger programs like Head Start that typically come from non-experimental research strategies, so comparisons of kids that attend Head Start to those who don't, and those also tend to find positive effects, but with that type of research design, you're always a bit worried, is there something else that's different between the Head Start kids and the non-Head Start kids, so our study is kind of bridging that gap. We have both a pretty big program in Boston serving a lot of kids and also this randomized design, kind of a natural experiment that comes from the assignment process that lets us get a clean estimate of what the program actually did to kids' outcomes.

Sarah Gonzalez: When we say preschool, what ages are we talking here?

Christopher Walters: Our study is about age four.

Sarah Gonzalez: What was just your big finding on the role of preschool in a person's life?

Christopher Walters: The main finding from this new study, which I should say is a joint project with Parag Pathak and Guthrie Gray-Lobe, my collaborators, the main finding from our study is that when you look 20 years out from preschool attendance using this randomized strategy, we find pretty large positive effects of attending preschool on college outcomes, in particular college attendance. Kids who randomly win a seat to attend a public preschool are about eight percentage points more likely to attend college on time, the year after they graduate high school, and six percentage points more likely to attend college at any time in the future.

Sarah Gonzalez: What about when it came to things like disciplinary matters, suspensions, getting in trouble? Did you look at that?

Christopher Walters: Yes, we looked at that too. One interesting finding in our study is we actually see nothing on test scores. We see no impacts on students like state achievement test scores, but we do see impacts on disciplinary outcomes like suspensions, truancy. We actually found a little something on reductions in juvenile incarceration, and when you put all that together, it's actually a pretty big decline in the rate at which kids are subject to various types of disciplinary actions.

Those two findings together on the test scores and the discipline suggest that the long-term outcomes we're finding are perhaps operating through a behavioral or socio-emotional channel rather than a test score, cognitive channel.

Sarah Gonzalez: The positive effects, were they different across racial or income groups or across sexes?

Christopher Walters: We looked at all of those dimensions. We didn't find any statistically significant differences by race or by income. The evidence there suggests improvements for all of those groups. By sex, we did find larger benefits for boys than for girls. Most of the outcomes that we saw, the disciplinary outcomes that you mentioned, college attendance, high school graduation, taking the SAT, all of those things improved more for boys in the program than for girls.

Sarah Gonzalez: They improved more for boys. Okay. What about the quality of the preschool education that these students received? Was it the same quality across the board at every school?

Christopher Walters: There's some variation across the programs in Boston. It's run on what's called an autonomous district model, so there's some latitude across schools to implement different curricula, hire different teachers. The program has actually, by some metrics, improved in quality since the time period of our study. One disadvantage of these long-term studies is we have to look way back in time, so we're kind of evaluating an old version of the program, and the evidence we have now suggests that Boston's program has actually gotten better since the time period of our study.

Sarah Gonzalez: Over the years, the education has gotten better, but comparing school to school, you're saying it tended to be similar in quality?

Christopher Walters: Yes. There's some variation, but we don't have much data on that, and what we're presenting in the study is essentially the overall impact of a typical program run in a public school district in Boston in the late '90s.

Sarah Gonzalez: You followed these children for years. What does that actually look like? When do you check in? What kinds of important milestones do you look at? Just explain to me a little bit about the process of something that span so much time.

Christopher Walters: A lot of this project has really just been waiting for the kids to get old enough so that we can measure the outcomes that we're most interested in. We've been working on this for like 12 years or so.

Sarah Gonzalez: Wow.

Christopher Walters: Essentially we waited until kids were old enough to attend college and then we were able to access this data source called the National Student Clearinghouse, which is kind of a big database that lets you check who's enrolled in college and where, and so we checked on all the applicants who applied to Boston public preschools in the late '90s and early 2000s and just measured which colleges did they go to, when did they enroll, how did they do, and compared those who were lotteried in to those who were lotteried out.

Sarah Gonzalez: You're following the kids who did make it into the lottery and the kids who did not. You sort of start at the kindergarten level I imagine for the other kids, the kids who did not make the lottery?

Christopher Walters: The interesting thing about our study is we can actually, regardless of whether where kids go to kindergarten, whether they leave Massachusetts, whether they go to a private school, we can take everybody who initially participated in the preschool lottery. Because we have records on their names and dates of birth, we can figure out where they went to college 20 years later, whether they attended kindergarten in Massachusetts or not. That's one advantage of our study, is we have pretty good follow-up rates on those sorts of outcomes, because we can follow everyone including those who didn't win the lottery.

Sarah Gonzalez: Okay. President Biden has proposed $200 billion towards free, universal preschool. How can your study help inform us on what that might look like?

Christopher Walters: That's a great question. As scientists, we're always a little bit hesitant to extrapolate from what we found in one setting to other settings. This program would certainly be broader than the Boston program we looked at in the study, but at the same time, building on the evidence that we discussed earlier on small-scale experiments and non-experimental preschool evaluations, this is kind of one of the first pieces of evidence I think we have from a randomized research design on a large-scale program that's really serving a large share of kids in a particular city and a universal program that's open to all children in the city.

In that sense, this is closer to what you might see if we did universal Pre-K, then the programs that have been evaluated in some previous work, and what we find is that, in Boston at least, this kind of universal program generated comparable benefits in terms of educational attainment, similar to what we saw on some of these smaller-scale studies earlier. At least from that perspective, it's kind of a possibility result and optimistic story for large-scale preschool, and we'd at least hope to see something similar if programs like this were implemented on a large scale elsewhere.

Sarah Gonzalez: Christopher Walters is an Associate Professor of Economics at UC Berkeley and is affiliated with the MIT School Effectiveness and Inequality Initiative. Chris, thank you so much for your time.

Christopher Walters: Thanks very much, Sarah.

Copyright © 2021 New York Public Radio. All rights reserved. Visit our website terms of use at www.wnyc.org for further information.

New York Public Radio transcripts are created on a rush deadline, often by contractors. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of New York Public Radio’s programming is the audio record.