collage of pictures of teachers and students in the shape of a head silhouette Accountability plays a critical role in today’s school systems

It’s just one of the many roles principals take on every day, but it certainly is crucial—teacher evaluation.

So, is there a single blueprint for teachers to follow? Hardly. In fact, teacher evaluation strategies have never been in short supply, and neither has the controversy surrounding them.

Jayne Ellspermann, principal at West Port High School in Ocala, FL, and president-elect of NASSP, says she has seen some fundamental changes in recent years. “Gone are the days where a teacher was not observed once by an administrator all year and then given an evaluation at the end of the year that said ‘meets expectations,'” she explains. “Well-implemented teacher evaluation systems can create a common understanding of what effective instruction is and support teachers as they work to help students succeed.”

Marc Tucker, president of the National Center on Education and the Economy, recently detailed a teacher evaluation plan, “Fixing Our National Accountability System,” which discusses strategies for holding schools—and teachers—accountable. The plan revolves around better testing, high-quality teachers, and a change in culture in the schools.

“No Child Left Behind radically shifted the balance of power in American education policy​making from the states to the federal government,” Tucker says. “It was not because a new consensus had emerged to make such a shift, but because both Democrats and Republicans were angry with the nation’s teachers, holding them responsible for a massive increase in the costs of our schools, while failing to deliver much in the way of improved student performance in return.”

“We got what we deserved,” Tucker says. And what schools got was a faulty structure of “gotcha” evaluations that didn’t really make anything better.

“Virtually everyone agrees that teacher evaluation in the United States needs an overhaul,” says Stanford University Professor and Researcher Linda Darling-Hammond in her report Creating a Comprehensive System for Evaluating and Supporting Effective Teaching. She also notes that existing systems “rarely help teachers improve or clearly distinguish those who are succeeding from those who are struggling.”

In another recent report, How Do Principals Really Improve Schools?, Rick DuFour and Mike Mattos, experienced former principals and now education consultants and authors, make a case for professional learning communities. However, they point out that mandated reforms “have consistently proven ineffective in raising student achievement.” The two cite research that they say shows 75 percent of teachers report their evaluation has no impact on their work and that “teacher evaluation does not recognize good teaching, leaves poor teaching unaddressed, and does not inform decision making in any meaningful way.”

Principals might agree, generally, but many also believe evaluations can improve teaching at their school. The main problem? Many evaluation systems require too much time and effort during a busy school day. Other principals are concerned about ever-changing mandates. Peter Fusaro, principal at Flathead High School in Kalispell, MT, and president of the Montana Association of Secondary School Principals, says good teaching is essential to successful education, but that requires accountability. “A great deal has been foisted upon us over the past seven to ten years,” Fusaro says. “This is my 22nd year in administration, and it is vastly different from when I started. However, one area that has not changed is the importance of teacher evaluations.” 

Multiple Measures

Generally, if there is one consistent theme in teacher evaluation, it’s that states and schools have shifted to require a mix of strategies. According to a 2013 report titled Trends in Teacher Evaluation about how states are measuring teacher performance, 41 states require that a combination of data be used to evaluate teachers, says Jim Hull, a senior policy analyst at the Center for Public Education who authored the report. 

“For most, this means evaluations take into consideration student achievement as measured by standardized scores, alongside traditional methods like classroom observations, lesson plan reviews, and others,” Hull says. “Combined, these measures make for a more accurate assessment while providing valuable feedback to teachers on their strengths and weaknesses.” Darling-Hammond says evaluations today “include multifaceted evidence of teacher practice, student learning, and professional contributions.”

Charles Pearson, superintendent of the Normandy School District near St. Louis, MO, says most states are now requiring student growth measures with multiple points of data over a two-year period for teachers, along with feedback models that connect “observed instructional behavior to specific learning outcomes.” Today, he says, “evaluation systems are developed from research-based instructional strategies, and data is collected on levels of student engagement and the level of rigor in lessons.”

The combination of methods must be well-conceived, says Robby Dodd, a former middle school principal and now a consulting principal in Montgomery County, MD. “A mixed approach that incorporates research-based practices clustered within several instructional and professional standards makes for a comprehensive evaluation system. However, such a system also needs to incorporate a way in which observers can pinpoint the impact of the teacher on student learning,” he says. “Otherwise, observation really becomes more about the way the teacher is ‘performing’ and less about the way he or she is instructing.”

Examples of Mixed Approaches

This mix of methods is taking shape in a variety of ways. In Maine, for instance, while a combination of new approaches is being tried, a narrative evaluation based on a preconference, a classroom visit, and a postconference had been the norm, says Marty Bouchard, principal of Houlton Middle/High School in Houlton, ME. “These evaluations happen with various frequency across the state and that has often been the case.”

Observations are generally a key ingredient in teacher evaluations (34 states require them), but they are taking different forms. “The feedback given to teachers is much improved,” Ellspermann says. “Most protocols have a rubric that is aligned with specific research-based expectations. Schools and school districts are specifying a number of formal and informal observations, and walkthroughs are now quantified to give teachers feedback.” They are more transparent, she adds, and because they have rubrics and expectations, teachers know what they are being evaluated on during an observation.

Hull says multiple observations performed by more than just administrators are the most effective. “Districts should consider training teams of observers; research has found that observations are more accurate when they are conducted by more than one person,” he says.

Measures of student achievement are a top priority, and a host of new methods of assessing students are always being recommended. “Instead of testing all of our students every year with low-quality tests, students should take high-quality accountability tests, covering a full core curriculum, only three times in their school career,” Tucker says. He wants higher-quality tests that reflect a better understanding of the material needed in careers and life today. Hull says measures of student achievement and student growth today are cheaper and easier to administer, more accurate for evaluating teachers, and better reflect student knowledge. Two models are generally found-value-added models designed to isolate the impact a teacher has on academic growth from other factors like their environment or previous scores, and student growth percentiles that measure student progress relative to others.

Darling-Hammond worries that popular value-added methods still are unfair, unstable, and affected significantly by the specific types of students a teacher has and by a host of factors that can be affecting those students. “The key is that the assessments be appropriate for the curriculum and the students being taught,” she says. “This may mean the use of tailored assessments for certain students.”

Robert Bobb, former president of the Washington, D.C., Board of Education and administrator in Detroit, MI, says he believes frustration with standardized testing has led to a significant lowering of standards. “Instead of ditching assessments, we should be discussing their quality. I believe the Partnership for Assessment of Readiness for College and Careers (PARCC), rolled out for the first time this spring in D.C. and 11 states, represents a radical shift from the standardized testing paradigm of the past,” he says. PARCC assesses skills most important to 21st-century success, such as clear communication, critical thinking, and problem solving. “[The PARCC system] provides parents and teachers meaningful feedback about students’ performance, crucial to building each child’s strengths and addressing specific areas of academic challenges,” he says.

In New Jersey, controversy arose over a plan to use PARCC over two years as 10 percent of a teacher’s observation, possibly growing to 20 percent. In other states, teachers and parents complained it was yet another test that would inaccurately measure student achievement and unfairly rank teachers.

Meanwhile, using student learning objectives, or SLOs, as an assessment tool has gained popularity and has been applied statewide in seven states, Hull reports, especially where standardized testing has been de-emphasized. His study suggests that SLOs reflect professional judgment, help evaluate the progress of individual students, and should be able to be applied to all teachers. However, he points out, they may not work broadly across a district. Another disadvantage to using SLOs is that they may cause teachers to set lower goals for some students, and they aren’t standard, which can make comparisons across districts or states difficult, Hull notes.

The U.S. Department of Education reported last year that 25 states had SLOs as part of their teacher evaluation process. DuFour and Mattos claim that data “unequivocally” shows student learning increases with professional learning communities because teachers are taking collective responsibility for student learning. Plus, they can share teaching practices and leadership.

Darling-Hammond stresses the need for common statewide standards that guide preparation, licensing, on-the-job evaluation, and ongoing professional learning. She also calls for a standard performance-based licensing assessment to assure that “teachers-in-training can plan, teach, and evaluate student learning effectively.”

Measuring Teacher Progress

There are various approaches to tying evaluations to teacher pay and position. Hull found that 19 states used evaluations for compensation levels, and 13 connected them with promotions. “Currently, most states/districts do not connect employment to student performance, but I believe this aspect of the evaluation system may develop in the near future, given our interpretation of the word ‘accountability,'” Pearson says.

Other ways that schools can measure teacher progress include:

  • Student/parent surveys 

  • Lesson plan reviews 

  • Teacher self-assessments 

  • Measures of professional learning

  • Student artifacts 

  • Teacher portfolios 

“These other measures can provide rich, qualitative information about classroom instruction,” Hull says. “As such, they add to the body of feedback teachers receive on their individual strengths and weaknesses. However, not much is known about how accurate these other tools are at measuring the effectiveness of a teacher’s practice. The exceptions are well-designed student surveys, which recent research shows are strongly correlated to results on value-added measures and classroom observations. 

“Whatever measures are used, districts should monitor their usefulness alongside other, more proven indicators to make sure that evaluations live up to their promise,” Hull explains. 

Finding Time

Bouchard is pleased to see important new approaches, but is wary that some are “so broad that administrators and teacher associations are choosing a small portion to pilot” even though teachers need high-quality feedback on all aspects of their work. One popular framework, for instance, offers four domains, and just the first one alone can be consuming, he says.

“If we take a narrow focus and just use that [single thing to evaluate staff], then teachers may go without feedback in other key areas for an extended period. For example, if my district only focused on classroom management, a teacher may not get high-quality feedback on important things like planning and preparation, which is domain four,” Bouchard says.

“Our school uses the Charlotte Danielson Framework for Teaching tool to provide feedback for our teachers,” says Brad Seamer, secondary principal at McCook Central School in Salem, SD, and president of the South Dakota Association of Secondary School Principals. “I love the process, but it is time-intensive, and it is difficult to do it justice in a normal school day. Doing evaluations and providing instructional feedback for my teachers is one of the things I enjoy most about my job. However, like most principals, I am stretched pretty thin, and I struggle to find time to be in the classrooms,” he says.

DuFour and Mattos see similar problems, citing an example in Tennessee (which was one of the first states to get an Obama administration Race to the Top grant). The model for teacher evaluation called for 50 percent of a teacher’s assessment to be based on a preconference, followed by principal observations (done four to six times a year in four areas, with 116 subcategories) plus a postconference; 35 percent was to be based on student growth; and 15 percent depended on student achievement data. Principals reported that each observation required four to six hours to complete. “No doubt these requirements are well-intentioned, but we’re convinced that advocates of this approach fail to recognize the crushing demands on the contemporary principal,” DuFour and Mattos write.

Hull says, “In most states, principals or their designee are responsible for conducting observations of all teachers. While staff evaluation is rightly a principal’s responsibility, finding time to conduct a comprehensive observation of all teachers, even once a year, can be difficult for these busy administrators.”

“The biggest challenge we still have in teacher evaluations is time,” agrees Ellspermann. “There is so much going on in our schools, and there is a limit to the time in the day. Time to get into classrooms and meet with teachers regarding their observations is hard to fit into the school day.” 

But several principals suggest this role is a critical function for school administrators. “While comprehensive teacher evaluation systems are vital to improving the profession and student outcomes, the role that principals play is equally important,” Dodd says. “If administrators aren’t regularly joining teacher teams during planning—frequently in classrooms observing and analyzing instruction, and working shoulder-to-shoulder with teachers to figure out who is learning and who isn’t—then the best-constructed system of teacher evaluation will have little impact,” he says. “The best teacher evaluation systems include deep and frequent examination of practice and make principals accountable for instructional leadership that supports teacher growth.”

Sidebar: Teacher Evaluation Systems: What Designers Should Ask

A recent report by the Center for Public Education, Trends in Teacher Evaluation, provided questions that designers of teacher evaluation systems should be asking, including: 

  • How much weight does each measure carry in the overall score? 

  • What measures are included in the evaluation system?

  • How accurate are the results?

  • Are the same measures used to evaluate all teachers? If not, how do they differ? 

  • How often are teachers observed in a classroom setting? Who conducts the observations?

  • How are teachers trained?

  • Is the observation rubric researched-based?

  • When do teachers receive feedback from each observation?

  • What other evidence of student learning will be used?

  • Are the overall evaluation results used to improve instructional quality? If so, how?

  • Are the results used for personnel decisions? If so, how?

  • Are the results made public? If so, what information is made public?

Jim Paterson is a writer, editor, and a school counselor and educational adviser in Silver Spring, MD.