By Jon Marcus New England Center for, Investigative Reporting
---- — Students at a privately operated online school that is costing Massachusetts taxpayers almost $2.5 million a year are falling far behind other students in the state, based on their assessment-test scores, and half of them are quitting during the academic year or failing to return the next year.
State and local records reviewed by the New England Center for Investigative Reporting show that the Massachusetts Virtual Academy, or MAVA, ranked second lowest statewide in its students’ progress in math and English based on a measure called the student growth percentile, which compares a given student’s MCAS scores over time with those of similar students.
Twenty-five percent dropped out last year, and, each fall, another 20 to 30 percent do not come back.
The results come at a time when legislators are considering allowing up to 10 online schools to operate across the state. The schools could enroll as many as 19,000 students.
A spin-off of the Greenfield Public Schools, MAVA accepts students from 148 other Massachusetts school districts including Lawrence, Lowell, Attleboro, Worcester, Boston, Fall River, Springfield, New Bedford and districts on Cape Cod.
Merrimack Valley districts with students enrolled in MAVA this school year include Methuen (8 students), Lawrence (5), Haverhill (4), Pentucket (2) and Amesbury (1).
The districts pay Greenfield $5,000 per year, per student. Greenfield, in turn, contracts with a Virginia-based company called K12 to provide instruction and other services.
The actual MCAS math scores for students at MAVA, now in its third year of operation, are lower than those of all but four other schools or districts, including a charter school for the arts in Gloucester and a charter school in Springfield that is on academic probation.
Thirty percent of the MAVA students received the lowest rating of “warning/failing” on the MCAS in math and 20 percent in science — double and almost double the state average.
“This is an atrocious change in the direction of public education,” said Maryelen Calderwood, a Greenfield School Committee member who has been an outspoken critic of the online school. “And they’re using my money and your money to make money on less-than-mediocre instruction. If it were up to me, I’d close it tomorrow.”
The next evolution in the school-choice movement, online schools in America are growing at a rate of 30 percent per year, according to the Center for Digital Education.
Virtual schools like MAVA, their advocates say, are designed primarily for students who have medical conditions that interfere with attendance, have been bullied or have other problems that make them hesitant to go to brick-and-mortar schools, are pregnant or parenting or are gifted.
“I still have deep regret we did not have this option in place in Massachusetts when a student opted to take her life rather than return to her school,” said Greenfield Superintendent Susan Hollins, a former New Hampshire charter-school consultant.
She was referring to the suicide of South Hadley student Phoebe Prince after she was bullied by classmates. “My first priority would have been a safe place to continue education, not the score on the state test.”
She also questioned whether the MCAS test is a fair measure of progress.
“We do not have students for years over time where we can be responsible for what they have learned before or how to take the test for best result, which many teachers go over with their students for months, or in some cases all year,” Hollins said.
K12’s director of academic analytics, Kerri Pickett-Hoffman, said students at MAVA are, in fact, doing better than national norms on another test the company uses to track their progress, called the Scantron Performance Series exam.
Those results, which the company made available, show that MAVA students made gains between October and May last year that were slightly higher than the national norm in math and much higher than the national norm in reading.
And while the students’ growth as measured by their MCAS scores is among the lowest in the state, it improved last year over the year before, said Jennifer Sims, K12’s regional vice president. “They are not as low as they were,” Sims said.
Still, between 2010-2011 and 2011-2012, the number of students at MAVA proficient in Grade 8 English and Grade 6 math on the MCAS test did not go up. It went down. The number failing math and science in Grade 8 increased, state records show. So did the number rated as needing improvement in fifth-grade math and science and seventh-grade math and English.
Sims said K12 has started a “national math lab” to help provide remedial math to students who need it. “We do have students who enter further behind in math, and we have to close that gap,” she said.
She and other school officials said it was misleading to compare the school’s test scores with those of entire districts. They said the results reflect the caliber of students who choose the virtual school. More than half, Hollins said, arrive performing below grade level in math and nearly 40 percent below grade level in reading.
“Most kids who are shopping for an educational alternative may not be the top-performing students,” said Picket-Hoffman. “They’re coming in behind.”
By several measures, however, those incoming students do not appear significantly different than the broader population.
Although K12, under its contract with the Greenfield district, is required to provide regular reports about its students, Hollins responded to a public-records request by saying that no document existed that gave specific information about the characteristics of entering students, enrollment or the number of withdrawals.
In fact, the report, obtained independently by the New England Center for Investigative Reporting, titled “Report to Superintendent Susan D. Hollins,” shows that just under 45 percent of the students were from families with incomes low enough to qualify for free or reduced-price lunches, higher than the statewide average of 35 percent.
But the proportion who required special education — 4 percent — was much lower than the statewide average of 17 percent. Fewer than one in five was enrolled because of a medical condition, fear of bullying or other safety issues and only one student because she was pregnant or parenting. And 28 percent were enrolled because they were taking advanced courses, not because they were behind.
Although the report does not specify what percentage of students arrived at the school below their grade level in math or reading, 44 percent were characterized as “unique learners.”
“The argument that these schools are enrolling a higher percentage of at-risk students, which is one of the things they claim, isn’t actually true,” said Michael Barbour, a professor of education at Wayne State University in Detroit who studies online learning. “They’re working with roughly the same kids as our regular schools, and they’re producing results that aren’t as good.”
The virtual school recruits its students by word of mouth and online, where K12 often shows up at the top of search-engine responses. Parents who make inquiries through the website are called within minutes by K12 representatives. Any Massachusetts student can enroll.
Legislation just passed by the state Senate and now pending a final vote in the House of Representatives would vastly expand the number of online schools that could operate in Massachusetts.
Partly in response to the poor results at MAVA, however, any future virtual schools would be put under state jurisdiction if they draw their enrollment from more than one district, said one of the bill’s principal backers, state Rep. Martha “Marty” Walz, D-Boston.
“We drafted it in part based on the experience with MAVA and in part on the experience in other states,” Walz said. “Somebody’s got to oversee the movement of kids from one district to another, and the transfer of money from one district to another.”
Walz said it’s also important for state officials to determine why so many students are leaving MAVA.
A report this summer by the National Education Policy Center at the University of Colorado urged that no more full-time virtual-education programs be approved until authorities understand why online schools’ performance is lower than those of other schools, and how it can be improved.
Mitchell Chester, state commissioner of elementary and secondary education, said Massachusetts should allow more online schools — though he added that the poor performance of MAVA “absolutely” demonstrates why they should be under stat control.
He said the way the virtual academy won approval leaves “no provision for the state to exercise either consumer protection or quality assurance with a school that effectively is a statewide institution.”
MAVA had three teachers, one for every 73 students, in 2010-2011, the last year for which the figure is available, compared to one for every 14 students statewide. Hollins said the MAVA staff has since grown to seven full-time and six part-time instructors. She said the average teacher salary is $40,000 — far lower than the state average of $70,340.
Students enrolled in the virtual school interact “a minimum average” of four times a week online with a teacher, Hollins said.
Each is otherwise supervised by a “learning coach,” usually a parent.
“Honestly, I did 99 percent of the work,” said Christina Refford, who enrolled two of her three children at MAVA but withdrew them eight weeks later; they’re now home-schooled. “We might go a whole week and never really hear from the teacher.”
Refford said she supports the idea of online schools. “But do I really want to see 10 more virtual schools opening up in the state? I don’t think so. If there were kids clamoring to get in, I would want to see more of it. But there aren’t.”
An internal school-district report prepared in March shows that 112 of the 472 students at MAVA last year, or about 25 percent, withdrew during the 2011-12 school year. The proportion of students statewide who withdraw from conventional schools is less than one in 20, or about 5 per cent. Hollins said another 20 to 30 percent do not return to the virtual school each fall.
It’s not clear what happens to the students who leave. Based on the experience of virtual schools in other states, experts said that some probably drop out altogether. Some return to their home districts, putting the burden on those schools to bring them up to speed.
Legislative audits and other reviews in Pennsylvania, Wisconsin, Arizona, Colorado, and Minnesota have shown that students tend to be further behind when they leave online schools than when they arrived, said Barbour.
Massachusetts school districts that have sent students to the virtual school are largely mute about what happens to them when they come back. Springfield, for example, which sends the largest number of students to MAVA of any single district, had 56 students enrolled last year, which fell to 45 this year. Springfield spokeswoman Azell Cavaan said the district would not comment on what happened to the rest.
Enrollment in online schools nationwide has been exploding. Many are privately run. K12, based in Herndon, Va., is the biggest provider.
The results in Massachusetts line up with K12’s performance elsewhere, according to Gary Miron, a professor at Western Michigan University’s College of Education and Human Development and coauthor of the National Education Policy Center report.
Students at K12-operated virtual schools lag behind other students in their states on assessment tests, Miron found, by between two and 11 percentage points in reading and 14 and 36 percentage points in math.
K12 last year made $522.4 million in revenues, up 36 percent over the year before, according to documents on file with the U.S. Securities and Exchange Commission. Enrollment in its schools rose nearly 46 percent, to 98,890 students.
Under the three-year contract approved by the Greenfield School Committee, which renews automatically, K12 ends up with 97 percent of the revenue from MAVA, the Greenfield School District gets the rest, and the two split any unspent money at the end of each year. That makes Greenfield’s annual cut about $69,000.
“They saw dollar signs,” Gary Aubin, who was chairman of Greenfield’s school committee before the contract was signed, said of the committee members who succeeded him. “But if you read the contract, the dollars go to K12, not to us.”
Rep. Walz largely paved the way for MAVA by adding a provision to the 2010 education reform law allowing Greenfield to open the school, although the state Department of Elementary and Secondary Education had at first objected. The department also blocked a subsequent proposal from the town of Hadley.
K12 paid $45,000 last year to the Franklin Street lobbying firm of Pierce Haley, and company executives have contributed at least $2,850 to Walz’s re-election campaigns since 2008, according to documents filed with the Secretary of State’s Office.
Asked about the contributions, Walz said the legislation she helped draft has both provisions K12 supports and provisions it opposes, including one that would ban for-profit companies from receiving future virtual-school contracts from the state — though they still would be able to provide instructional services as subcontractors.
“My focus is on educational opportunities that virtual schools offer and the approval and accountability measures that strengthen the state’s oversight of virtual schools,” Walz said.
The New England Center for Investigative Reporting (www.necir-bu.org) is a nonprofit investigative reporting newsroom based at Boston University and supported in pat by media outletsthat include The Eagle-Tribune.