A weekend interview about Florida's voucher program with Northwestern U. researcher David Figlio
Florida has the largest education voucher system in the country. And up to now, no one has been able to quantify whether children who use the vouchers to go to private schools get better schooling than those who stay behind. Northwestern University professor David Figlio finally has put some data to the question, and has concluded -- at least for now -- that those who leave perform no better or worse than those who don't. He spoke with reporter Jeff Solochek about his research, his next steps, and his views on the spin that voucher advocates and opponents have given to his work.
What exactly prompted you to take a look at the way the voucher results are turning out?
Well, I've had a very longstanding interest in both school choice in general as well as the education of disadvantaged kids. This is the largest school voucher program in the United States, so in that regard I think it's a question of considerable national interest. And given that the vouchers are aimed specifically at low-income kids -- or kids from low-income families, that is -- it's targeting a key target population that is of direct interest to me. So to me it was very natural for me to think about what is happening to these schools, who is using the voucher programs, what schools are they choosing to go to, and is the voucher program leading to increased competition with the public schools, and are the kids doing better in the voucher program. Some of these questions I've investigated, and others I'm in the process of investigating now.
Were you surprised with what you discovered so far?
Not really. Well, okay. I'll tell you one thing that might be a little bit of a surprise to some people, although it wasn't much of a surprise to me because I expected it could go either way, is one result. The kids that participate in the corporate tax credit scholarship program tend to be worse off than other kids. They tend to be the ones really struggling. They're struggling kids going to struggling schools. They tend to be lower income. ... A lot of people who have historically opposed vouchers have done so because they argue it will lead to cream skimming, where the most motivated, the best-off kids, the highest performing ones would be the ones who leave. And in Florida, in this program, it looks like that is not true. It's the lowest-performing kids that are leaving disproportionately to use a voucher. ...
Did you talk to any of the participating families about why that might be the case?
Well, we're actually in the process now. Actually, in the spring I conducted a pretty large-scale survey of 700 families and we're in the process of coding up those data now. ... We asked questions not only about the satisfaction with the schools. And this is for people who chose to use the voucher and for those who were eligible to use a voucher but didn't use one. We were interested in what information they used when trying to pick their schools. Why they chose to participate in the program ... and how satisfied they were with their kids' schools. That's something I don't have the answers for that, but that's something for later on this summer. ...
You said that you felt the results you found weren't going to change too much. That the students who left didn't really outperform those who stayed behind. What made you so confident in those results?
Well, okay. I think confident may be too strong a word. I believe my quote was that I have a hunch that this might be the case. And the main reason I have the hunch is ... there are very few studies out there that show that kids in one sector do dramatically better than kids being educated in another sector. You just look at the distribution of gains in the data, and the kids participating in the program on average tend to be doing as well as the kids who weren't participating in the program. There are selection problems still. Mainly that choosing to participate in the program is not a random thing. ... And there's a lot of missing data from 06-07. So 07-08 is the first year I feel confident with the data, the private test score data. But with those provisos, the selection issues would have to be monumental to generate very large differences between the two. ...
What do you think that people should be taking away from what you found? Because people seem to be taking all sorts of things away from what you found.
Well, I'll tell you, what I personally think one should do is not rush to judgment either way. I've seen a fair amount of spin on both sides of the voucher debate already. One spin, and I guess this is the spin of your employer's editorial page at least, is, Hey, look, unless kids are doing a lot better in the voucher program than in the public schools, then the voucher program is not a success and we ought to scotch it. Another spin which I've seen by more voucher advocates is, Hey, look, if the kids in the program are doing just as well and their education is costing taxpayers less, then that means the private schools are moer efficient and so we ought to expand the voucher program.
I reject both of those conclusions based on this year's data. Okay? And the reason that I reject both of those conclusions is that I feel we need to have stronger causal evidence on the relative effectiveness of the program. And I don't think we can get the stronger causal evidence now, until I have two years of really strong private test score data. I think that's a necessary but not sufficient condition for strong causal evidence. But in order for me to drill down and deal with some of these selection stories, I really need to have confidence that I am working with a complete data set.