Students at the Avalon School in St. Paul, Minn., take part in a holiday gift exchange in 2010. (Elizabeth Flores/Minneapolis Star Tribune)

David Osborne, co-author of “Reinventing Government” and other books on public-sector reform, directs the Progressive Policy Institute’s education reform project.

Walk through a typical public school, and you see students, sitting in rows of identical desks, listening to teachers talk. Unless the teacher is particularly inspiring, half of the students are zoning out. This isn’t just a problem for teachers, half of whom leave the profession within their first five years. It’s also a problem for their pupils: Disengaged teenagers do not make the best students.

Now imagine if students were instead encouraged to work on projects they chose: building robots, writing plays, researching why bees are dying off by the millions.

When teachers run their own schools, they often make such changes. “We’re competing against Xbox 360, and over-scheduled days with soccer practices and very dynamic lives,” says Kartal Jaquette, one of 10 teachers who run the Denver Green School. “Are you almost as interesting as a video game? Are you getting almost as much attention as a soccer coach might? Is it as much fun? Because if not, they’re going to tune you out.”

Teachers are in charge of at least 70 public schools in 15 states; most, but not all, are charter schools. Ten more teacher-run schools, including one in Maryland’s Prince George’s County, are in the planning stages. These schools are not only redesigning the learning process to better engage students, they’re improving student performance. On top of that, they’re stemming the high dropout rate among teachers.

Studies show that the average teacher reaches maximum effectiveness after about five years in the classroom. When nearly half of all teachers leave the profession within five years, we are losing talent we desperately need.

There are many reasons for this high dropout rate. But in my years researching education, the complaint I’ve heard from teachers most often is: “They treat us like children.” Polls bear this out: Last spring, Gallup reported that of 12 professions, teachers were the least likely to agree that “at work, my opinions seem to count.”

Most teachers have no say in their schools’ decisions about hiring, promotions, firing, budgets, pay levels, curriculum or scheduling. This lack of control is a big reason they leave the profession, according to the University of Pennsylvania’s Richard Ingersoll, who studies teacher retention.

By contrast, consider the Avalon School, a charter middle and high school in St. Paul, Minn., which opened in 2001. A converted warehouse with high ceilings and exposed pipes, Avalon looks more like a loft space for artists than a public school. Two lead teachers share most of the school’s administrative duties, but all decisions — curriculum, schedule, salaries — are made by the entire group of 28. They meet two mornings a week, marching through each agenda item in a matter of minutes.

“I have a lot of friends in more traditional models,” says Tim Quealy, who teaches math, technology and language arts at Avalon. “They are just told what to do — some big binder lands on their desk, and their days are scripted. They feel very isolated.”

Avalon has committees that handle specific duties: personnel, technology, special education. Every year teachers evaluate one another on each other on four questions: What are their contributions? What are their greatest strengths and skills? What is some constructive feedback? And how confident are you in their overall performance? Parents and students also evaluate teachers, using different questions. If problems surface, the personnel committee appoints a fellow teacher to mentor his or her struggling colleague. If that fails, the group lets the teacher go, which appears to happen more often when teachers are in charge than it does in traditional public schools.

Having more control keeps teachers and students more engaged. Avalon’s high schoolers can take math, biology, physics and Spanish classes, but they spend the majority of their time on projects of their own choosing, with guidance from teachers to ensure that they master state standards. Such a heavy reliance on independent projects is typical of teacher-run schools, according to Kim Farris-Berg and Edward Dirkswager, who studied 11 of them for their 2012 book, “Trusting Teachers With School Success: What Happens When Teachers Call the Shots .”

When I spoke with Avalon students, it was obvious they were immersed in their own educations. One girl I met had adapted a book into a play and directed it when she was in ninth grade. Another wrote an interactive murder mystery and produced it with a classmate, raising $200 for their prom through ticket sales. A senior boy cooked six meals from different periods of American history. While the teachers ate, he explained the context for each one: that one consisted entirely of beans because it was for slaves, why there was no salt in another meal and so on.

When projects don’t cover all related state standards, teachers — who work more as coaches than instructors — intervene. One student, concerned by the mass die-offs of bees, did a project on bees rather than take biology. He researched threats to bees; visited beekeepers, apiaries and a state bee lab; and listened to TED talks on the subject. Once he was done, his teacher, Jo Sullivan, identified the state standards on water and carbon cycles he had missed, asked him to research them, then required him to demonstrate mastery.

Every sophomore and junior must do a major project. And to graduate, seniors must complete a 300-hour project, working with an expert from the wider community, and present it to the entire school. “This model is empowering to both the student, because they get to pick what they learn about, and to the teachers,” Sullivan says.

Teacher retention, on a year-to-year basis, averages 95 percent at Avalon, according to lead teacher Carrie Bakken. That’s higher than typical rates in St. Paul district schools, and about 10 percentage points higher than the national average in urban schools.

Charter schools in St. Paul get 24 percent less money per child than district schools, and 40 percent of Avalon’s students have a learning disability. (The project-based approach is well-suited for special education.) Still, Avalon outperforms the St. Paul average on most standardized tests and the state average on some. And its teachers value other measures more, such as the quality of senior projects. In a survey of about 125 graduates, 74 percent were in a post-secondary program or had completed one, and 88 percent agreed that their senior project had helped prepare them.

There are many different teacher-run models; some schools have principals, but teachers make the key decisions, even selecting the principal. Denver Green School (DGS), an “innovation school” with charter-like autonomy that opened in 2010, is a teacher partnership, organized much like a law or consulting firm.

Three lead partners spend most of their time on administration, while still teaching one class a day. But decisions are made by all 10 partners. In addition to fairly full teaching loads, the other seven partners each take charge of one area, such as hiring or professional development. The other 32 teachers are employees who can join the partnership if invited.

Though its students are younger than Avalon’s — pre-K through eighth grade — DGS also relies on student projects more than most other schools. Students tend an organic garden on campus, which provides about 80 pounds of food a week to the school cafeteria from August through October. The day before I visited, the sixth grade put on a harvest festival, during which they reaped the produce and offered organic foods from the garden. While preparing, they read a teenage version of Michael Pollan’s “The Omnivore’s Dilemma,” plus a book about a Cleveland urban landscape that was turned into an organic garden.

All the project work is connected to their classwork. In science class, for instance, the sixth-graders had been studying the water cycle, from rain and snow falling in the Rockies to rivers running out to the sea and evaporating. They then evaluated the use of drip irrigation in the garden and determined that it saved 1 million gallons of water a year, compared with when the building was vacant and the district was watering weeds.

One year, teacher Kartal Jaquette’s second-grade class counted every light in the building, as part of a math project to find out where energy was being wasted. They measured the lumens coming from each light, as well as from mini-skylights, or solar tubes. Using graphs and charts, they figured out which lights they could unscrew. Then they designed a monitoring system, with a student “light sheriff” to make sure that every classroom had enough light but didn’t waste energy. Their recommendations, according to lead partner Frank Coyne, saved $1,200 and 20,000 tons of carbon dioxide a year.

Though DGS prioritizes project-based learning over test prep, its academic results are strong. In ratings of academic achievement at all Denver schools, compiled by the district, DGS is in the second-highest of five categories, “meets expectations.” Teachers who consistently weave projects into their lessons produce some of the school’s highest test scores. Last year, 26 percent of Jaquette’s third-graders tested proficient and 60 percent tested advanced in math. In a school where 60 percent of students are low-income, half are minorities and 27 percent are English-language learners, that is a home run. “There’s not another school with that demographic [in the city] who had that level” of proficiency, he adds.

The advocacy group Education Evolving, which just published a guide to creating teacher-run schools, released a poll of teachers and members of the public last year that illustrates why the idea is spreading. After hearing a description of teacher-run schools, 78 percent of teachers surveyed liked the idea. More than half of non-teacher respondents were “very interested” in seeing one in their community, and one in five teachers wanted to implement the idea immediately. Interestingly, those sentiments didn’t change among union members.

The biggest obstacles to the spread of teacher-run schools are school districts’ central rules, most of which make it impossible to use unusual personnel configurations, alter budgets and make myriad other changes the teacher-run model demands. That’s why so many teacher-run schools are charters — they need autonomy to organize as they please.

Many union leaders love the teacher-run model as much as they hate charters. They constantly argue that teachers should be treated as professionals, and there is no more professional model than a teacher-run school. In Minnesota, in fact, the Federation of Teachers has created an organization to authorize teacher-run charters. In that state, and perhaps in others, this model might carve out some islands of truce in the war between unions and charters.

More important, in an era of resistance to tax increases, most districts can’t solve their teacher-retention problems by raising salaries. Handing teachers more control is probably our best shot at keeping more quality teachers in the classroom.

Read more from Outlook and follow our updates on Facebook and Twitter.