Teacher Evaluations: Presumption of Incompetence

(Image from Wiki Commons)

In San Francisco Unified (SFUSD) last year, 40 out of 1,924 teachers (2%) received bad reviews. Over the past five years, an average of 2.7% teachers received marks of “unsatisfactory” or “needs improvement.” In nearby San Jose and Oakland, only around 1% of teachers received poor ratings.

Abuse of Data Leads To Abuse Of Teachers

The Ed Deformers love to hold up numbers like these as proof that the teacher evaluation system is broken. What the Ed Deformers are actually demonstrating by such abuse of data is their own ignorance and incompetence. Low numbers of unsatisfactory reviews does not prove that the evaluation system is broken any more than high temperatures indicate a broken thermometer. These results could mean that there are actually very few bad teachers. There is no reason to assume that there must be more bad teachers than are being caught by the current system or that there is necessarily a bell curve of teacher aptitude.

On the contrary; many evaluation systems, like that used in California, are based on benchmarks or standards that teachers are expected to work toward. Teacher training programs, student teaching experiences and professional development are generally geared toward these standards. Teachers are assessed on these standards yearly until they have achieved tenure, and every two to four years thereafter. In the assessment process, teachers can receive evaluations of unsatisfactory, satisfactory or better. Obviously, unsatisfactory implies that some standards have not been met, while satisfactory means that they have. While some teachers are certainly better than others at any given standard, a rating of “satisfactory” must still be accepted for what it is: the ability to sufficiently perform one’s job. Thus, contrary to all the anti-teacher rhetoric, we should expect the vast majority of teachers be able to meet these standards if they have been adequately trained, as most have.

That this is true should be clear by examining a sampling of the California teaching standards:

• 1.4 The teacher actively engages all students in problem solving and critical thinking within and across subject areas
• 2.2 The teacher maintains a safe learning environment in which all students are treated fairly and respectfully
• 2.4 The teacher ensures that expectations for student behavior are established early, consistently maintained, and clearly understood
• 3.1 The teacher demonstrates knowledge of subject matter content and students’ cognitive development
• 5.1 The teacher establishes and communicates learning goals for all students
• 5.4 The teacher uses the results of assessments to guide instruction
• 6.3 The teacher collaborates with faculty and staff

Each of these standards is clearly an important part of good teaching (as are the others I left out for the sake of space). They are relatively easy to learn, both through training and practice, and to demonstrate to administrators who are evaluating teachers. Therefore, it should not be surprising that the vast majority of teachers get positive revues. It is not only what we should expect for evaluations based on standards (rather than test scores), but what we should desire for our children.

Excellent Teaching Does Not Always Produce Excellent Students

On the other hand, receiving a satisfactory to excellent review, indeed being an excellent teacher, cannot guarantee that every student is learning the student content standards. This is because students come to school with a range of prerequisite skills and social development, familial support structures, and material security (or lack thereof) that influence their academic success. Therefore, it is entirely possible, indeed likely, that we can have good teachers with bad student test scores. This may be part of the reason for the irrational assumption that there must be many more bad teachers out there than are caught under current evaluation systems.

Evaluation Systems are Limited by Their Evaluators

Another problem is that administrators are stretched thin with all their other responsibilities, which means that they seldom have the time for many classroom observations of their teachers or to give adequate attention to teacher portfolios. This may result in some teachers receiving satisfactory evaluations when in fact they actually need improvement.

Irrational Reforms Can’t Solve Imaginary Problems

Oakland Superintendent Anthony Smith said “We have to create a better evaluation system that really names what high-quality instruction looks like,” but we already have that in California. The Standards for the Teaching Professional clearly identify the qualities of good teaching. His solution to include student performance in teacher evaluations, in contrast, is irrational and cannot possibly help to achieve his goal. Student test scores tell us nothing about what good teaching should look like. Student achievement data is at best a proxy that may or may not tell us anything about the quality of teaching. For example, a good student may do well on a test, even when the teaching is very poor, because that student already knew how to study, and had considerable background knowledge, high self-efficacy and the ability to focus for extended periods of time. Student achievement data, unlike teaching standards, tell us nothing about methods to motivate students, or how to reach low income students, English language learners or special needs students. In contrast, while benchmark or standards-based evaluations do not guarantee improved test scores, they can and generally do describe what good teaching should look like and can help teachers reflect upon their teaching and improve.