A plan to strengthen K-12 education, released earlier this week, sets many ambitious deadlines. By the end of the 2012 legislative session for example, Education Commissioner Stephen Bowen wants lawmakers to agree on tough expectations that Maine teachers must meet to be considered effective in the classroom. He's also calling for statewide standards for evaluation programs, in large part, to guide local school districts, as they develop their own systems for grading their teachers performance. But figuring out exactly what a tough, but fair, teacher evaluation should look like isn't always easy.

It's a challenge that started to get more attention three years ago, when the federal government set up this competition called Race to the Top. States could vie for more than $4 billion in education money from the Recovery Act. But they had to submit applications, laying out in exacting detail their strategies for doing everything from turning around failing schools to improving classroom teaching through better training and evaluations.

Maine's application was rejected. "And the place where we got hammered in the scoring was around that teacher and leader piece," says Stephen Bowen, the state's Education Commissioner.

Winning states had some innovative ideas on training and evaluation. But many states, like Maine, were told they had a long way to go. Bowen says Race to the Top helped start a conversation. "States all over the country and the feds are looking at what works--effective classroom practices and using new types of evaluation systems and data systems," he says.

Before Race to the Top, the U.S. was more or less in the wilderness on teacher training and evaluation. Nothing showed this as dramatically as a landmark 2009 study called The Widget Effect. The report, put out by The New Teacher Project in Washington D.C., looked at evaluation and dismissal policies in four states and 12 school districts.

Of the more than 15,000 teachers who took part in the research, less than one percent received unsatisfactory ratings, even if their schools were consistently failing. And roughly three in four teachers got no feedback on how to improve in their most recent performance review.

"We're trying to move to systems that will tell us who our highest achieving teachers are, tell us who our true chronic underperformers are and tell us much more about everybody in between," says Sandi Jacobs, vice president of the National Council on Teacher Quality. "Figuring out the exact nuts and bolts of the systems that are going to get us there is definitely hard work. And it's a heavy lift."

A school board meeting Thursday night in Chelsea, near Augusta, showed just how heavy. Greg Potter is Superintendent in Sheepscot Valley Regional School District.

"So the scorecard you see, the template itself is set up so that it can be expanded," Potter says as he shows members of the school board a mock-up of a report card on an overhead projector. This district, and four others in Maine, are piloting evaluation and training programs, courtesy of a grant from the federal goverment.

The report card grades teachers on student growth in reading and math, predicted proficiency in both and average daily classroom attendance.

"When you talk about average daily attendance, are you talking about students or teacher?" asks board member Barbara Skehan. "Student attendance," Potter says.

That doesn't sit well with Skehan, who's a teacher herself at nearby Cony High School in Augusta. "Unless you're going to give me keys to a car to go and pick up a student, some of which have lower attendance, and unless you have the authority to go and get that kid, you're going to dock me," she says.

Other board members followed Skehan with questions about other aspects of the evaluation model. They're the kind frank conversations that are likely to go on across the state in the coming months and years, as policymakers, legislators, school distirct administrators and teachers try to come up with new approachs to teacher evaluation that are tough, but fair.