Seth Zimmerman is a Ph.D. candidate in the Department of Economics who has conducted research on the effect of school choice on test score performance and attendance in low-income urban schools. Zimmerman co-authored a working paper for the National Bureau of Economic Research titled “The Effect of School Choice on Intrinsic Motivation and Academic Outcomes,” which found a 21-percent decline in truancy rates for males entering high school after they found out they would be attending their first choice school. He spoke to the News about the implications of his current research.
Q What’s the essence of school choice?
A School choice is when students have the option to attend schools other than those in their local school zones. In the district we studied, there is a pretty comprehensive district-wide choice program where students can choose from a variety of public magnet schools or charter schools. Each year there is a lottery in the spring where students will submit a ranked list of the schools they want to attend. After the district conducts the lottery the students find out where they are going to go the next year.
Q Do you think that the motivation to go to a choice school would in itself have an effect on lowered truancy rates and improved test scores?
A I think that in general if you were comparing students who wanted to attend a new school or attended a school of choice to students who didn’t, that would be a problem. What we do in the paper to deal with that is we use these lotteries. We say, anyone who wants to attend a school, signs up for the lottery. Then, within the lottery, some students are randomly selected to have that opportunity. If you compare the students who wanted to go versus the students that wanted to go but weren’t selected, that is a pretty clean comparison because they would both in theory have the same level of motivation.
Q But couldn’t the students who applied for but were not accepted to the lottery program also affect truancy rate if, let’s say, they lost motivation due to the rejection?
A What we do to think about that is we compare the students who win the lottery to the students who lose the lottery, and we then compare both groups to students who don’t enter at all. The idea is the students who don’t enter at all don’t receive a positive or negative shock, and what you see is the students who lose the lottery and the students who don’t enter at all basically continue along the same trend, whereas the students who win the lottery tend to have lower truancy rates. That’s sort of the way we deal with that.
Q How do you think your study would fare in terms of repeatability and scalability?
A There have been other studies of these high-performing charter schools, especially in Boston and neighboring cities that have come up with results that are qualitatively similar to ours in terms of finding these large test score effects. I would say that the result is beginning to appear more frequently. There have also been other large-scale studies on the effect of school choice, like in New York City … , that have found modest but positive effects. I would say, in general, our test score findings are broadly consistent with current research.
Q In the paper you focused on truancy rate change after students were admitted to their choice school, but before they actually enrolled. What was the reasoning behind that?
A The idea is that students that get to attend their chosen school benefit in at least two ways. On the one hand, they may benefit if their new school is better than the school they might have attended, or if the school is a better match for them. That could be true for how they score on tests, or for how they behave in the classroom if the pedagogical style is better suited to them. At the same time, students may benefit if they are more motivated and bring more to the table themselves when they have the opportunity to attend a school that they want to attend. Now, the problem is that generally those two things take place at the same time, so you are becoming more motivated, but you are also attending a school that may be a better match for you. The insight that we had is that there is a brief period of time when students haven’t been directly exposed to this new teaching that happens at their chosen school, but may already have accrued some of the motivational benefits just by knowing they have the opportunity to go in the future.
Q That kind of reminds me of college admissions.
A Yeah, right? College admissions might be the opposite though because once you find out you got in, it’s over [laughs].
Q Are you continuing this research?
A One project I’m currently involved with is a project I am doing with one of the co-authors of this paper, Chris Neilson, about school construction. New Haven has one of the biggest per capita school construction projects in the country and it’s been sort of a high priority of the mayor and the school administration for about 20 years now. They’ve rebuilt nearly every school in the city. If you walk around New Haven you’ll notice that a lot of these elementary schools are really beautiful. [Cooperative Arts and Humanities High School] downtown was rebuilt under this project. [Worthington] Hooker School at East Rock was rebuilt as well. It’s about a $1.5 billion project. Chris and I worked with the New Haven district to evaluate that project and tried to understand what impact rebuilding the schools had on the students in terms of scores and also what the impact was on neighborhoods and home prices around the schools. [We found] that these schools had relatively large and sustained effects on student performance and the neighborhoods that surround the schools had increases in home prices. If they build a new school in your school zone, the value of the surrounding homes will tend to go up by a modest to large amount, I would say. This suggests that people value these projects because people are willing to pay more for a house that has default access to the newly constructed schools.
Q What about your upcoming projects?
A We’re doing a big project in Chile right now where we’re trying to run a large informational experiment with the goal of seeing how having better information about the costs and benefits of different college degree programs affect students’ choices about where to attend college. Chile is facing many of the similar issues that we are facing here in the sense that they have a partially privatized higher education system together with public loans. [This means that] choices that students make about where to go to college have consequences as far as student loan default, [which can pose] big costs to the government. In some ways it is a more extreme situation than we have here.