The number of K-12 students enrolling in full-time virtual and blended learning schools continues to grow, despite research suggesting that students in these programs do not perform as well as their peers in traditional settings.

In a series of new research briefs released Tuesday, the nonprofit National Education Policy Center (NEPC) challenges the idea that online learning programs can be as effective as traditional schools in delivering individualized instruction to students. And it shows that, for all the flexibility they offer, such programs often come at a cost to student performance.

Enrollment in virtual learning programs is still growing, though not quite as rapidly as it has in recent years, says Michael Barbour, an associate professor of instructional design at Touro University California and a fellow at the NEPC who co-authored the research briefs. In the 2017-18 school year, nearly 300,000 students were enrolled across 501 full-time virtual schools, and an additional 132,000 students were enrolled in 300 blended learning schools—an increase of 2,000 and 16,000 students over the previous year, respectively.

But even as the sector grows, one thing remains constant, Barbour says: “Students in these programs—both full-time online programs and blended schools—tend not to do as well as their brick-and-mortar counterparts.”

He adds: “There’s not really a rationale for the growth, based on performance.”

According to the three-part briefs from the NEPC, a research organization based out of the University of Colorado at Boulder, graduation rates at virtual schools and blended learning schools hover around 50 percent and 61 percent, respectively, while the national average for public schools is 85 percent.

Under the broad umbrella of virtual schools, which operate in 39 U.S. states, other notable trends emerge: 56 percent of district-operated virtual schools earned “acceptable” performance ratings based on the researchers’ review of data provided by state education agencies, compared to 40 percent of charter-operated ones. And independent virtual schools perform better (59 percent) than those operated by nonprofit education management organizations (50 percent), and both perform better than those run by for-profit EMOs (29 percent).

Additionally, the report finds that virtual schools tend to enroll fewer students of color and low-income students (those who qualify for free and reduced lunch).

One of the elements that’s lacking in the virtual school sector, the NEPC authors argue, is accountability. But according to Barbour, policymakers “haven’t had an appetite” for introducing legislation that would provide oversight of virtual, online and blended learning programs, except for in a few rare cases where “something dramatic” has already happened, as with Ohio’s Electronic Classroom of Tomorrow (ECOT). In that case, the online charter school had been inflating the number of students it enrolled in order to receive additional state funding, and as a result, the school was forced to close in 2018.

“In the last couple of years, we’ve seen more bills introduced on oversight, but they’re not going anywhere,” Barbour says, adding that the bills typically die before they’re even taken to a vote. This, he says, could be the result of lobbying by companies that stand to gain from a growing virtual school sector. It could also be because virtual learning programs demand lower operating costs than a traditional school.

Since 2013, the NEPC has regularly issued reports on the policy, performance and research around virtual schools in the U.S. Every year the conclusion is largely the same: students in these online learning programs are not as successful as those in traditional brick-and-mortar schools.

Each report also offers suggestions for improving the state of the industry. This year’s edition recommends that, until achievement rates begin to improve, policymakers slow or altogether stop the growth of virtual schools, sanction low-performing schools, sponsor additional research on the effectiveness of virtual schools and require that student-teacher ratios be lowered.

The current student-teacher ratio in virtual schools is 2.7 times higher than in U.S. public schools, the NEPC researchers found. On average, national public school classrooms have 16 students for every teacher, compared to 44 in virtual schools and 34 in blended learning schools.

Reducing the student-teacher ratio in virtual and blended schools, Barbour says, won’t have an instant effect on student performance, but the smaller class sizes will allow teachers to “engage in several different pedagogical strategies that are easier to employ with a lower number of students.” As is, teachers have so many students that they can’t provide the personalized learning experiences that so many online learning programs promise, he adds.

“The reality is the type of interaction occurring [between students and teachers] in these automated settings is cursory and requires a low level of student knowledge,” he says. “What you are essentially getting is the type of interaction and feedback that can be provided through multiple-choice questioning.”

Some virtual schools, Barbour acknowledges, are “bucking the trend” and posting acceptable achievement levels, but he says those schools tend to be exceptions to the rule.