Kindergartners in Georgia — many of whom don’t yet read — could soon play an important role in deciding which teachers get raises or get fired. Under a new pilot program, 5-year-olds will be guided through a survey that includes such statements as “My teacher knows a lot about what he or she teaches” and “My teacher gives me help when I need it.” As the youngsters circle a smiley face, a neutral face or a frowning face, they will be playing their part in new high-stakes teacher evaluations.

The kindergartners could help put Georgia at the forefront of a growing national movement to make student surveys part of how teachers are rated. Students in every grade across the state will participate in the pilot program, and, depending on its results, Georgia may incorporate the student feedback into teacher evaluations as early as next school year, when it will join such measures as observations by principals and student test scores. The state has yet to determine how much weight the student evaluations will carry in teacher ratings.

Although Georgia is the only state so far to consider using students to grade teachers, individual school systems from Washoe County, Nev., to Pittsburgh are launching similar pilot projects. Memphis already counts student survey results as 5 percent of a teacher’s overall evaluation. By the fall of 2013, that figure will be 10 percent in the Chicago public schools.

“The idea that student feedback can tell us something meaningful about classroom instruction is exceptionally important,” said Tim Daly, president of TNTP, a New York-based organization formerly known as The New Teacher Project. “The notion of taking direct input from students is a major development in evaluating teachers.”

D.C. officials have drawn wide attention for including a test score growth metric known as “value-added” in teacher evaluations. Student feedback, though, is not part of the formula.

“While we have piloted student surveys in a few of our schools this year solely as a means of giving teachers feedback on their performance, and plan to make them more widely available next year, we do not currently include them as part of our teacher evaluation system and have no immediate plans to do so,” said Jason Kamras, chief of human capital for D.C. schools.

Kids can be ‘quite biased’

Teacher evaluations have traditionally involved principals or other administrators observing a teacher once every few years. In many school systems, nearly all teachers earned satisfactory ratings, according to a TNTP report in 2009. But in light of research demonstrating that individual teachers are the most important in-school factor influencing student achievement, researchers and education officials nationwide are searching for the best way to identify teachers who need extra help.

Some research also suggests there is a relationship between student performance and what students say about teachers. But questions remain about how best to use these surveys — and how young is too young for students to take them.

Rob Ramsdell, a director of the Tripod Project, which has been designing and administering student surveys since the late 1990s, advised caution. To Ramsdell, the point of student surveys is to give teachers more information about what is — and isn’t — working in the classroom. “There probably is a place for them in teacher evaluation systems, but we think the use in that way needs to be handled very carefully,” he said.

Some Georgia teachers are also skeptical. Lena Nwakudu, a high school special-education teacher in the Emanuel County schools in central Georgia, said she didn’t know much about what the student surveys will look like, but doubts they will be useful. “I’m not in agreement with it, unless it’s really monitored for specific things it’s asking for,” said Nwakudu, president of the local teachers union. “Kids, number one, don’t have the maturity to do it and, number two, can be quite biased.”

‘Not a popularity contest’

Questions in student surveys are framed to focus more on the day-to-day operations of a classroom, providing evidence of good teaching strategies instead of a student’s opinion of a teacher, proponents say. For example, instead of asking a student whether he or she likes a teacher, a survey might ask whether a teacher reviews material at the end of every class or how well classmates behave.

“It’s not a popularity contest,” Ramsdell said. “We’re asking questions [teachers] would actually like to know the answers to, getting at teaching practices that they fully appreciate are important.”

Nina Esposito-Visgitis, president of the Pittsburgh Federation of Teachers, said she is wary of plans to include the Tripod surveys in teacher evaluations because of administrative problems in the city’s two-year-old pilot program. In some cases, she said, students have even been linked to the wrong teacher.

Pittsburgh officials acknowledged that some problems had cropped up, but said they were working to fix them before survey results are used in evaluations.

“They have to administer it with the care of a high-stakes test — because it is,” Esposito-Visgitis said. “If this is going to be part of a teacher evaluation, it’s high-stakes.” She also said she would like to see more research on how well surveys work.

Tripod Project surveys, as well as those Georgia will implement, are designed to be given to students from kindergarten through high school. Ryan Balch, a doctoral student at Vanderbilt University, developed and administered one in a seven-district pilot program in Georgia last year. He limited his surveys to grades four and higher.

“To be honest, I don’t think it’s feasible in most situations with students in kindergarten through second grade,” he said. Balch added that there are questions about surveying very young students who can’t yet read and may have subjective feelings about a teacher.

Officials in Georgia and elsewhere working with student surveys said they would wait for results from the pilot programs before deciding which grades to include.

Ramsdell is confident, though, that Tripod surveys can offer useful feedback at any age level. The surveys, he said, show “very consistent patterns that lead us to think that we’re picking up information that would be helpful feedback for teachers.”

Pittsburgh has administered Tripod surveys for the past two years, but has kept the results confidential, even from most teachers. Now, like other districts, it’s working toward incorporating the results into a teacher evaluation formula that will also include student test scores and principal observations.

Student surveys “can actually give you real feedback about where you can get better and how you can improve,” said Samuel Franklin, director of teacher effectiveness for the Pittsburgh schools. “It makes sense when you think about it. Students are in the classroom every day. They are experts on what’s happening in the classroom.”

This story was produced by The Hechinger Report, a nonprofit, nonpartisan education news outlet based at Teachers College, Columbia University.