Attending a sought-after charter middle school didn’t increase a student’s chance of attending or graduating college, a new U.S. Department of Education study showed.

The report, released Monday, also found little connection between charter school quality, as measured by test scores, and college outcomes.

“The overall conclusion that there is little difference between charter schools and non-charter schools is not shocking to me,” said Sarah Cohodes, a professor at Columbia University’s Teachers College. She pointed to prior research showing charters perform comparably to district schools nearby.

Cohodes said that while the study’s use of random lotteries allowed it to convincingly establish cause and effect, it looked at a relatively small sample of 31 schools, only three of which served predominantly low-income students.

Still, the results are a disappointing data point for charter advocates who hoped the publicly funded, privately run schools would improve students’ college prospects.

The research arm of the Department of Education released the study. A spokesperson for Secretary of Education Betsy DeVos, a proponent of charter schools, did not respond to an email seeking comment.

The analysis focuses on about 1,700 students who won a random lottery to attend one of these charter schools during the 2005-06 and 2006-07 school years; their outcomes were compared to students who applied to the same school but were not offered a seat.

As it turns out, the two groups fared roughly the same. Among lottery winners and losers, 69% would go on to attend college, and they attained college degrees at similar rates.

The new study follows previous research on the same set of middle schools, which found these schools didn’t lead to higher test scores. It also showed that these schools had no effect on rates of suspension or absenteeism, but they did improve parents’ and students’ satisfaction with their school.

In that study, test scores varied by student group, with low-income kids seeing gains from attending a charter and higher-income students seeing declines. But the new study revealed that attending a charter had no effect, positive or negative, on the college prospects of either low-income students or their more affluent peers.

Notably, charter schools that raised test scores did not necessarily boost college enrollment — and vice versa. This bolsters arguments that schools’ effects on math and reading test scores aren’t good indicators of longer-term success.

“Federal and state policies promoting school improvement often presume a link between improving students’ academic achievement and their later success in life,” wrote study authors Kate Place and Philip Gleason, of the research firm Mathematica. “Yet several recent studies are beginning to question that assumption.”

The research on this issue remains somewhat mixed.

“It is an interesting data point which lines up with some other research that says things that impact test scores don’t necessarily impact long-term outcomes and vice versa, but it is only one data point in what I hope are more data points to come,” said Cohodes.

It may also be harder for middle schools, as opposed to high schools, to make a difference in students’ college trajectories.

“If students who attended the charter middle schools systematically sorted into lower performing high schools then the effect of those high schools could be misinterpreted as the effect of the middle schools,” Nathan Barrett, senior director of research and evaluation at the National Alliance for Public Charter Schools, said. “The study does not explore this possibility.”.

A key caveat of the paper is it focuses on a relatively small sample of 31 charter school sites from over a decade ago. That’s less than 10 percent of all charter middle schools operating at the time; last school year there were over 7,000 charter schools across the country.

The average school in the study also served significantly fewer low-income students than charter schools writ large, either at the time or currently. That’s notable because the charters with the strongest track record at raising test scores are so-called “no excuses” charters, which predominantly serve low-income students of color.

“This study was not designed to look at the impacts of ‘no-excuses’ style charters and I don’t think it should be interpreted as telling us information about those schools,” said Cohodes, who said she’s hoping further research will connect different charter practices to student outcomes.

Other research in a handful of states, including a recently published study in Texas, suggests charter school performance, as measured by test scores, has improved over time.

The use of lottery data limits which schools can be included. The researchers were only able to get lottery data from a small share of charter middle schools, in part because many of them were able to enroll all those who applied. If charter schools that have higher demand are more effective, as has been shown in some research, then the study may actually be an overly optimistic look at the impact of charter schools more broadly.

“This study focused on a limited number of charter middle schools, and all of the schools in the study held admissions lotteries,” the study authors say. “The results might have been different if the study included a different or larger set of charter schools.”

The latest national study adds to a small body of research on longer-term outcomes for charter school students.

A separate study found that Texas charter schools overall led to small increases in college enrollment, but modest declines in income as adults. “No-excuses” charter schools like KIPP and IDEA led to fairly large increases in test scores, high school graduation, and four-year college enrollment and persistence, though had no clear effect on income.

Studies in Chicago, Florida, and Georgia have found that attending a charter high school increases students’ persistence in college.