Lots of research on charter schools amounts to a horse race: Which schools produce better test scores or higher college enrollment rates, district schools or charter schools?

But that doesn’t say anything about how charter schools affect the broader community — whether they spur surrounding public schools to improve or hurt their performance by draining resources.

It’s “the most important question in the charter school debate: How does charter growth affect student outcomes system wide?” said Marty West, a professor at Harvard. “That is a very difficult question to answer.”

That’s why two recent studies are significant. They are among only a handful that attempt to isolate how the arrival of charter schools affects the achievement of all students in an area, whether they attend a charter or not.

Both offer evidence that charter schools lead to net gains in student test scores, at least for some students. But the case is not yet closed — and we still don’t know much about the other ways charter growth affects students. Here’s what the studies tell us.

As charters grow in cities, black student performance rises, according to a new study.

A study released last week by the Fordham Institute, a pro-charter think tank, used national data to show that charter school expansion is linked to slightly higher average reading test scores in grades 3-8. Math scores, though, didn’t rise.

(The study was funded by the Walton Family Foundation, which supports charter schools; Walton is also a funder of Chalkbeat.)

Black students in urban areas saw particularly large gains. For those students, living in a place that goes from zero to 50 percent of students attending charters was connected to test scores increasing by about a third of a grade level in small cities, and jumping nearly two-thirds of a grade level in large cities. There was some inconsistent evidence of gains for Hispanic students, and no clear effects on white students.

The results are generally in line with past research, which has shown that charter schools don’t clearly outperform district schools on average, though charters in cities serving mostly black and Hispanic students tend to do especially well.

But the Fordham Institute’s promotion of the results have raised eyebrows. “If you want to boost the achievement of kids of color, get charter enrollment to at least 50% in every big U.S. city. This will measurably move the needle — for everyone!” tweeted Fordham President Michael Petrilli Monday.

“It is a quantum leap to argue that moving to 50% charter enrollment will produce these gains,” countered Brendan Bartanen, a professor at Texas A&M University.

Petrilli partially walked back his comments Tuesday. “I could have been more careful with my tweets,” he wrote. “Still, there’s a clear relationship between market share and achievement.”

You should know: The findings need to be interpreted cautiously.

Charter schools aren’t randomly assigned to communities in a way that would allow researchers to easily track their impact. So the study uses complicated statistical methods to try to isolate how charters affect test scores.

Essentially, Fordham’s David Griffith looked at the test scores in grades 3-8 within each school district, using national data from 2009 to 2015. He compares growth in grades with more charter schools to the growth in grades with fewer charter schools, controlling for other factors like student poverty. These trends across many places yield an overall estimate of what happens to test scores of all students in an area — charter and district alike — as charter schools grow.

West, who offered input on the paper, called that a reasonable, albeit imperfect, approach.

The results do come with a number of significant caveats, some of which are laid out in the paper. For instance, the data does not allow Griffith to see whether higher-achieving students might have moved into the places with more charters. “We definitely pushed the limits of the data,” he said.

And the study also doesn’t examine how other factors might have helped schools improve in places that have seen charters grow dramatically, like spending increases. Charter schools in cities like New Orleans and Newark have been magnets for philanthropic dollars.

It’s not entirely clear, then, whether places that don’t have a lot of charter schools would see similar gains from expanding them, especially if the uncommonly successful charter operators don’t move into those communities. There is wide variation in test scores across different charter schools.

“High charter enrollment might be indicative of high-quality charters,” said Bartanen. “But unless those schools can be magically translated to any context, there is no guarantee that expanding charters everywhere will produce these gains.”

Some other research supports the idea that charter school growth leads to (small) gains in test scores. But other consequences haven’t been fully examined.

A separate, recently released study looking at North Carolina examines what happens to test scores in a given community after charter schools were allowed to grow due to a 2011 statewide policy change. This paper has more detailed student data than the Fordham study.

The average student in areas where charters grew saw their math scores increase, relative to similar students in places that didn’t see charter growth. The effect was quite small, and was driven by improvements in district schools. (The study does not break down results by student race.)

A draft study from 2017 looking at New York City found something similar. Charter schools there raised test scores overall, though they also increasingly isolated disadvantaged students in district schools.

These studies back up the general conclusion of the Fordham report, but they find gains that are quite small. And in most places, charter school enrollment is unlikely to grow at the rapid clip that the Fordham suggests would be necessary to see those large achievement gains.

Meanwhile, charter critics note that the schools can draw students and therefore resources away from district schools, leading to painful staffing cuts and even school closures. Financial analyses have backed this up, though the latest research suggests that doesn’t translate into worse test scores. But critics note that the consequences of those closures and staffing reductions — including disruption for students and the loss of community institutions or beloved teachers — go beyond academic outcomes.