RALEIGH – My son Alex and North Carolina’s charter-school movement are roughly the same age – a little over seven years old. Both have shown tremendous growth. Both show flashes of brilliance (proud daddies get to say that) and periods in which they stumble awkwardly.
There’s another commonality between charter schools and the Little Conqueror: the achievements and prosperity of both depend to a large degree on the efforts of parents.
The issue of parental choice in education has attracted significant debate for decades. In an odd way, however, the parental part often gets lost. There are dueling philosophies of public education, dueling assertions about the value of competition and markets, hosts of articles and studies looking at spending and achievement. But if advocates of choice are correct, then the ultimate test of educational success or failure lies with the evaluations of parents – parents who don’t necessarily share the same goals, values, situations, or respect for standardized tests.
That’s a general point to keep in mind when assessing the validity of new studies purporting to show that charters aren’t delivering the educational goods. A couple of weeks ago, the nation’s second-largest teacher union, the American Federation of Teachers, came out with a report that employed National Assessment of Educational Progress scores in an attempt to show that district-run public schools outperformed charters. More recently, and more relevantly, Duke University professor Helen Ladd and her co-author, Robert Bifulco, have published a paper that looks at North Carolina’s charter schools and concludes that students would be better off staying in district-run schools.
The AFT analysis is quite easy to debunk, and many already have. It makes no serious effort to distinguish the value added by each type of schooling, thus suffering from selection bias and other problems. By its logic, AFT should agree that private education is far superior to public education, because average scores in the former routinely exceed those in the latter.
The Ladd paper deserves to be taken much more seriously. It does attempt to measure the real value-added by district schools and charter schools, tracking individual students as they leave one for another. As far as education research goes, it is far superior to much of the nonsense that gets peddled as scientific or hyped as gospel. Indeed, when I read a draft of the paper some time ago, I thought it significant enough to deserve closer, scholarly attention. N.C. State professor Craig Newmark is currently examining the same data, seeking to replicate or explain Ladd’s results.
In the meantime, however, I have some thoughts. Not to get too technical about it, but at several points Ladd and Bifulco had choices to make about which measurements to use and which variables to include. In general, their choices might result in a bias against charter schools. For example, transferring students in and of itself often results in weaker performance in that year. Even if students will prosper in the long run, the change is disruptive and imposes costs. One potential advantage of charters, all other things being equal, is that many of them combine elementary grades with middle- and high-school grades– which means that their students have the opportunity to avoid costly transfers. Meanwhile, public-school districts often redistrict students for a variety of reasons. But these effects are not modeled in the Ladd/Bifulco paper.
Another problem is the diversity of charter schools. Many of them are targeted to particular groups, such as students at risk of dropping out. While Ladd and Bifulco acknowledge that it would be worthwhile to distinguish those charters from charters with more traditional academic goals, they did not do so. The data available now make it possible to do so and it should be done. Poor performance on the state tests by a few of these targeted charters could well mean that the authors’ results reveal little about the value of the great majority of charter schools.
The largest problem here, however, is that to allege that charter schools perform significantly worse than district schools do is inherently to question the choices made by the parents of the 20,000 charter-school students (as of 2003). In effect, the authors’ results suggest that 20,000 shoppers can indeed be wrong, either because parents don’t have the necessary information or because they are simply not qualified to make sound choices.
This argument is just a big question-begging exercise. Who decides what the right information is, or how to evaluate the choice? Some charters satisfy parents with higher scores on other, national tests more aligned to their curricula than the state’s tests are. Some satisfy parents with better discipline, safety, or convenience. Many satisfy parents in ways that are not quantifiable. Because educators constantly complain about not being treated as real professionals, it’s weird to see them insist on not being judged by the number of satisfied customers – which is how other professionals and businesses are evaluated every day.
Hood is president of the John Locke Foundation and publisher of Carolina Journal.