Those of us at Fordham have strived over the course of our organization’s two-decade existence to stay open to new evidence and to be willing to change our minds. For example, we shifted from the notion of "letting a thousand flowers bloom" when it came to charter schools to acknowledging that "some weeding is necessary" after multiple studies showed just how poor the achievement of some charters was turning out to be, and just how hard it was to actually shut such failing charter schools down. And there have been other smaller shifts over the years, too, on funding, teacher diversity, and more.
Of course, we are fortunate to be an independent think tank, with our own endowment and a mission to follow the evidence wherever it leads. It's not so easy to keep an open mind when you're an interest group, like a teachers union, which is charged with protecting its members' concerns. The unions are never going to say, "You know, we've changed our minds based on the evidence, and we’ve decided we really like these non-unionized charter schools." So for them the game is about finding evidence that supports their position and ignoring, if not discrediting, the rest.
We see that on our side of the reform fence too, as when some of our friends decided that test scores weren't valid measures of student success after multiple rigorous studies showed negative test score impacts of private school choice programs. Thus the energetic effort to discredit test scores as predictors of long term outcomes.
But back to reform opponents. There used to be a fierce debate about whether kids in charter schools were learning more than kids in traditional public schools. But now that dozens of rigorous studies have found that kids of color in urban charter schools learn significantly more on average than their district peers, the unions and other opponents have had to find other arguments to make their case, while working to discredit the impact evaluations.
So they now have a two-part argument.
First, they argue that the charter school advantage is due entirely to “creaming.” For example, in a recent Wall Street Journal article, the co-chairman of United Teachers Los Angeles attributed the success of charter schools to “having classes filled with motivated, high-performing students.” It's apparently not enough to claim that some of the advantage comes from selection—the likelihood that families who choose charters are different in important ways from those who don't. But all of it?
Second, they argue that, regardless of how good or bad charters might be for the kids they serve, their growth is hurting traditional public schools and the kids who are left behind there. This argument has the benefit of polling extremely well, and has been used to great—or terrible—effect in the current California charter school wars.
With that long context in mind, Rising Tide: Charter School Market Share and Student Achievement, Fordham’s new, highly significant analysis by senior research and policy associate, David Griffith, examines the relationship between charter school market share and student achievement—not just for the kids in charters or the kids in district schools but for everybody.
It sounds straightforward, but to our knowledge it is the first time anyone has conducted such a study.
We have the CREDO evaluations and other studies looking at the performance of students in charter versus district schools. And there have been many studies of “competitive” or “spillover” effects of charter schools on district schools, most of which find that competition from charter schools does not harm achievement in nearby district schools, and sometimes boosts it.
Rising Tide doesn't look at the differences between charter and district kids. In fact, we can't distinguish between them because our data source, the Stanford Education Data Archive (SEDA), includes the academic progress that all students in a given geographic community made compared to students in other geographic communities nationwide. In other words, the academic performance of charter schools is included in what SEDA refers to as the “geographic school district,” regardless of whether the charter schools operate independently of the district in which they are located (although we do know what percentage of students attended charters).
Griffith spent an entire year getting acquainted with the dataset and examining the best way to model “charter market share.” For example, because we’re really interested in the achievement of specific racial subgroups, it makes more sense to consider the effects of “charter market share” within those subgroups than across them. So his findings focus on the relationship between the percentage of black, white, or Hispanic students who enrolled in charters and the average achievement of all black, white, or Hispanic students in a geographic school district, including those in traditional public schools.
In the end, what we learned was quite simple, and quite powerful: For large urban districts, the more black and Hispanic students are enrolled in charter schools, the greater achievement is for black and Hispanic students.
This has huge implications for the two arguments that charter opponents are making.
First, it provides new evidence that creaming can't explain the entirety of the charter school advantage in urban districts. Because if charter schools' success was truly an "illusion," as the Wall Street Journal author charges, we'd see no gains for communities with greater charter market share. Higher test scores in charters would be canceled out by lower test scores in district schools, driven by the transfer of higher-achieving students from district to charter schools. It would be a zero-sum game.
Instead, we find achievement gains in districts with more charters. What that implies is that the additional learning that's happening in charter schools is not coming at the expense of less learning in district schools. It's additive. And that implies that kids in charters really are learning more—not because of who the kids are, but because of what the schools are doing.
The findings also have implications for opponents’ argument that charter schools are hurting traditional public schools. We suspected that wasn’t the case, based on existing evidence, at least for the district schools located closest to new charter schools, at least when it comes to test scores. But maybe the performance of other students in the district was being harmed in some way. We don't see any evidence of that in this study.
Of course, this one analysis won't end the charter school wars. Perhaps some scholars or bystanders who mildly opposed charter schools will change their minds, now that there's even more evidence that they really are doing something to help black and Hispanic kids learn more, and that they aren't hurting the outcomes of students in district schools. Yet we suspect that most critics will continue to oppose charter schools because their opposition has always been based on bread and butter interests, like the bargaining power of teachers unions, rather than evidence or reason.
But, charter supporters, stand proud. These schools really are getting better results for children of color, and not just because they are attracting the most motivated families.
That's good news with which to kick off the new school year.