Most of the time, teachers substantially work alone. How can anyone tell whether they are any good at what they do?

California law requires schools to "evaluate and assess" teachers, but in most cases the feedback teachers get is of very limited value, if any. Every once in a while, the principal or a designated evaluator walks by with a clipboard, watches for a few minutes from the back of the room, makes a few marks on a form, and leaves. During the Pandemic, many schools dispensed with even this pretense.

Extensive observation of a teacher is often a sign of trouble. Scrutiny is uncomfortable, and sometimes it is used as a way to put pressure on the teacher to improve, or even to quit. Some teacher contracts limit the number of times a principal may observe a teacher, or set rules that require the principal to provide advance notice for observation.

As discussed in Lesson 3.5, most school districts invest heavily in education for teachers — but only through the structure of the pay system, which rarely involves any kind of evaluative signals. Most districts pay extra for teachers that have earned advanced degrees, but spend very little to provide their faculty with ongoing training. Are the professional development programs they provide effective? The only way to know would be through evaluation. Evaluation of teachers is not an area of strength for the education system, especially in California.

Many states require teachers' evaluations to include students' results. But not California.

In 2009, the New Teacher Project wrote an important report criticizing the weakness of teacher evaluation processes. In a nod to economist Adam Smith, it was titled “The Widget Effect.” The report argues that “school systems treat all teachers as interchangeable parts, not professionals. Excellence goes unrecognized and poor performance goes unaddressed. This indifference to performance disrespects teachers and gambles with students’ lives.” The report contributed to a wave of policy actions.

To align the incentives of the education system toward stronger teaching, in the recovery plan for the Great Recession the Obama administration offered states funds to overhaul their education systems. The program reqired that states include student test score improvement as an element of teachers’ evaluations. Many states agreed to this requirement, but California declined.

In California, districts are not required to consider student test scores when evaluating teachers.

Many states require objective evidence of student achievement (usually test results) to be included as a factor when evaluating the effectiveness of a teacher. In the peak year of 2015, 43 states had this requirement. In 17 states the requirement was a bit more sophisticated; they required the evaluation not just to be based on test scores (which correlate with family income) but on individual growth of test scores for the students under a teacher's watch. In 2019, the National Council on Teacher Quality (NCTQ) updated its analysis in a detailed report comparing teacher evaluation practices and policies across the US states. 

Should student learning results have any influence on the evaluation of teachers? It's easy to agree that they should, but this principle has not found its way into policy in California.

In 2015 an education advocacy group filed a lawsuit (Doe v. Antioch et. al.) to push the issue. Pointing out that California law obligates school districts to evaluate teacher performance, they argued that scores on tests based on the state's standards should be part of the evaluation. The lawsuit failed to achieve its purpose. In 2016 Superior Court judge Barry Goode ruled that the law in question (the Stull Act) is not specific; it requires some kind of evaluation, but leaves the manner and consequences of evaluation up to school districts. Nothing changed.

Should teachers review each other?

An alternative approach, called Peer Assistance and Review (PAR) supports principals in some California districts in the task of evaluating teachers. In this system, districts invest in more frequent observation and evaluation, and try to make the process beneficial for the teachers being observed. Underperforming teachers are assigned a coach and evaluated by a teacher panel. There is some evidence that this approach is effective in raising teacher performance. It also may be helpful in “coaching out” some teachers who might do better in a different line of work. If managed carefully, PAR can help provide the required documentation to support a formal dismissal when called for. Critics of PAR express concern that it can have the opposite effect, creating hurdles and obstacles to dealing with performance issues in a clear, effective way.

In California, few districts have implemented PAR or other systems of constructive evaluation for teachers in part because they cost money. When it's time to negotiate a contract, most teacher unions have preferred to advocate for education dollars to go toward salaries rather than support for evaluations.

Why not ask the students?

Of course, the people in a school who are best-positioned to know a teacher's strengths and weaknesses are the ones carrying backpacks: the students. According to research by NCTQ, as of 2019, student feedback is a required element of teacher evaluations in seven states (AL, HI, IA, MA, MS, UT). In New York it is prohibited. In the rest of the states, including California, NCTQ says that student evaluation of teachers is either allowed or not mentioned in state law.

As test scores have fallen out of favor, some schools and districts have experimented with student feedback as a way to understand teacher strengths and weaknesses. The 2010 National Teacher of the Year, Sarah Brown Wessling, uses a version of the Tripod survey (see image below) to gather student feedback. Her version of the survey also includes space for students to write their own response.

California teachers can "opt out" of being evaluated by students.

In 2010 the California Association of Student Councils (CASC), a statewide organization of student leaders, argued that evaluation for teachers should include feedback from students. The student organization enthusiastically advocated for legislation to facilitate student involvement in teacher evaluation and celebrated passage of legislation that appeared to pave the way. No funding was provided for such evaluations, however, and student participation in teacher feedback remains rare. Ironically, the amended legislation that eventually passed appears to prohibit districts from making student evaluations a required practice: instead, it guarantees each teacher an unlimited option to opt out. This is a good example of how advocacy can backfire.

Another alternative that has been proposed but not yet tried (readers, please add a comment if you know otherwise) is for higher-grade teachers to evaluate lower-grade teachers based on the preparedness and work of the students they teach.

What should be done about "bad" teachers?

The demand for meaningful teacher evaluation systems gained urgency during the Great Recession. Why? Because school communities wanted a better answer for a tough question: when the budget requires laying off teachers, who should be the first to go? 

A 2013 PACE/Rossier survey of California voters on what idea would have the most positive impact on public schools found significant popular support for "removing bad teachers from the classroom" (43 percent), followed by "more involvement from parents" (33 percent), and "more money for school districts and schools" (25 percent)." Despite this popular suppor, the idea is difficult to into practice because it implies certainty about who, exactly, falls in that list of worst teachers. Test scores are generally lowest in places where poverty is greatest, and it can be challenging to untangle effectiveness from circumstance.

Surveys of voters, parents and teachers tend to agree that the system is failing to take action when teachers are ineffective; inaction on this issue became one of the core arguments in the 2014 case Vergara vs. California.

"Value Added" evaluation of teachers

Value added analysis is used in many fields in order to statistically tease out the likely contributing factors that lead to an outcome. This kind of analysis is incredibly powerful because it can help focus attention on individual performance and cut through the "noise" of things beyond control. It is used in everything from finance to baseball.

In analysis of education, the outcomes of interest are things like test scores, attendance, grades or graduation. Factors of interest in such a model can include things like family wealth, community conditions, school programs and, importantly, the teacher. Given enough data, a value-added analysis of patterns of success in a school can statistically indicate which specific teachers are consistently associated with desireable results or undesireable results. Value-added analysis statistically examines whether students are scoring as expected, given their circumstances, or differently. Over time, if a teacher's students tend to come out of their classes with more improvement in their scores than the model predicts, the teacher might be doing something worth celebrating. Of course, the reverse is also true.

The methodology for value-added analysis in education is steadily improving, and there is a lot to know. Harvard Professor Raj Chetty studies this topic, and his explanations are useful.

A positive view of evaluation for teachers

How can anyone become better at their work in the absence of meaningful feedback?

Teacher evaluation should be about much more than dismissal decisions. After all, how can anyone become better at their work in the absence of meaningful feedback? Greatness By Design, a California task force report on how to support outstanding teachers, stresses that educators should be evaluated against professional standards and that evaluation should be informed by data from a variety of sources, including measures of educator practice and student learning and growth.

Teachers don't necessarily reject evaluation if it helps them improve. This conclusion has been echoed by large-scale surveys of teachers by the Gates Foundation, which in 2009 began the MET project, an effort to develop useful evaluation models for teachers. In the video below from TED Talks Education with John Legend, Bill Gates argues for significant investment in support of teacher development.

One evaluation method that the MET research found to be particularly valid was student surveys.

Many pioneering schools and districts took inspiration from Charlotte Danielson's Professional Practice framework, which defines a rubric for evaluating and coaching teachers in order to make evaluations more consistent and focused.

Perhaps the most thoroughly developed, tested, respected (and also misused) system for teacher evaluation and improvement is Washington, D.C.'s IMPACT program. It has been in use and under development for over a decade. Partly because evaluations are connected to bonuses, the system has undergone intense scrutiny, revision and replication. If your district wants to consider an evaluation system for teachers, deeply understanding the IMPACT program is essential.

As testing has become an important component of school management and accountability, student test results have become a correspondingly important component of teacher evaluation in most states — but not in California. The topic continues to be a subject of debate.

Updated September 2018, October 2021.