As part of the 10 Ideas: A Millennial Lesson for Education series, a call to improve evaluations so that they better reflect teachers' skills while holding them accountable.
If you were asked to choose the teacher that had the biggest impact on you, how would you make your decision? Would you base it on his engaging teaching methods, the academic improvement you made under his tutelage, or the positive atmosphere of learning he created in the classroom every day? Education policymakers around the country are trying to tackle a similar question: How should we evaluate and compensate teachers fairly?
When it comes to evaluating teachers, America's educators have tended to focus on the inputs involved in the education process. These inputs include teacher education level, teacher experience, and the size of the classroom, among other things. Although an emphasis on inputs makes sense intuitively, recent studies (such as the above) have shown that they actually don't bear strong correlations with student achievement. There are some nuances to this, such as the significant improvement in teacher quality in the first several years of teaching experience. But these studies as a whole have shown what some people might know intuitively: just because a teacher has a master's degree and experience does not mean that she will be good at her job. This fact has led many researchers and policymakers to argue that teacher compensation should instead be at least partly outcome-based, or based on student achievement.
In fact, even in an increasingly polarized political realm, many policymakers on both sides of the aisle seem to agree that outcome-based methods of teacher evaluation should be an important component of teacher evaluations. The specific mechanism that many policymakers have chosen to use as the basis is the value-added model of measuring student achievement. These models use standardized testing to assess students' baseline level of knowledge. They are then able to calculate the value added to student achievement scores by teachers after controlling for variables such as socioeconomic status and race.
On the surface, the incorporation of these value-added measures into teacher evaluations (alongside principal evaluations, portfolio reviews, and more) seems ideal. After all, these value-added models ostensibly allow school districts to sort teachers based on effectiveness, reward the good teachers, and develop or replace the ineffective teachers.
In reality, however, value-added models are not categorically beneficial and should be incorporated with caution. The key reason that value-added models can be ineffective is that their standard errors can be worryingly high due to the low sample size involved. In fact, in discussing the instability of the distribution of students' test scores over the course of several grades, researchers at the Value-Added Research Center at the University of Wisconsin-Madison wrote, "the bottom line is that there is substantial inconsistency across grades, years, and cohorts in the means and standard deviations of Wisconsin test scores." For this reason, policymakers who employ value-added models should exercise an awareness of the potential problems involved and proceed with corresponding caution.
In addition to the currently excessive focus on the inputs of education, teacher evaluations (involving qualitative assessments by school principals) are also often flawed. In Georgia public schools, for instance, less than 1 percent of teachers are rated as performing their jobs in an unsatisfactory manner, although the true figure of unsatisfactory teachers who ought to be developed or replaced is significantly higher. Principal evaluations may be flawed because they have personal connections with the teachers, because they have no control over the teacher's salary or assignment, or because they are afraid of a potential lawsuit in the event that they do fire a teacher. In any case, principals appear unwilling to give teachers accurate qualitative evaluations, and this inhibits the school system's ability to deal with teachers effectively.
To solve this problem of poor qualitative evaluations by principals, I have suggested that the Georgia General Assembly use a portion of its allotted Race to the Top grant funding to create an independent body of teacher evaluators to provide unbiased qualitative assessments of teachers. This body of evaluators would be part of the Georgia Department of Education, and it would begin as a small pilot group to evaluate teachers using established principles of high quality teaching. The evaluators' methods would involve classroom observation, informal student interviews, and possibly more. The evaluators themselves would ideally be former teachers with many years of experience and who demonstrate high teaching ability. Each evaluator would only assess and provide comprehensive feedback to teachers in his or her area of expertise; former music teachers would evaluate music teachers, former middle school science teachers would evaluate middle school science teachers, and so on. Importantly, the body of evaluators would also have the funds and wherewithal to deal with potential lawsuits brought by teachers who are fired or replaced.
By incorporating the assessments of this independent body of evaluators into each teacher's assessment, Georgia public schools will be able to make more informed decisions regarding the compensation and development of teachers. This policy could also act as a middle road between the reformers who want to evaluate teachers with value-added models and the educators who feel that their work cannot be described by a standardized test administered to their students. By proceeding cautiously and evaluating teachers fairly, we ensure that our schools' most important resource is utilized properly.
Seth Taylor is a member of the Roosevelt Institute | Campus Network's chapter at the University of Georgia, where he is studying history and political science.