As the debate surrounding American education reform continues, more states are adopting a system of teacher evaluation that relies heavily (up to 50%) on their students’ standardized test scores. Considering the published studies showing correlations between increased test scores and student success in college and later in life, it’s not surprising that the Education Department has been encouraging school districts to adopt this new evaluation system. Eduardo Porter recently wrote an article for the Atlantic about these new education policies, and their potential for failure due to Goodhart’s law. Goodhart’s law was named after British economist Charles Goodhart, and explains how performance statistics (like a standardized test) are only valuable as performance statistics if they aren’t performance statistics. That might sound confusing, but it’s actually very straightforward: when you heavily weigh one aspect of a task or goal, people will do better in that one aspect (by whatever means necessary). This can be seen in every sector, be it hospitals that try to keep patients alive past Medicare’s 30-day “survival” measurement, or the Chicago PD lowering crime rates by simply classifying some arrests and busts as “noncriminal”.
Placing a large amount of a teacher’s worth and pay on standardized test scores makes it more likely that the teacher will spend less time teaching the students what they need to know, and more time preparing them for standardized testing. When the evaluation system focuses on score improvement, rather than the scores themselves, gifted students and students that are farther behind might be put on the back burner so teachers can focus on the students with potential to improve. Goodhart’s law makes this cause for concern. There are really only two ways to improve and meet a goal: put in the work to improve or cheat the system. When a teacher’s job or pay is put on the line, cheating become a more viable option. Schools across the country have been accused of cheating on test scores, unfairly dividing attention given to students and writing their curriculums around the tests.
Those against the new evaluation system argue standardized test scores are only a small part of teaching, and placing a teacher’s value on student’s test performance ignores everything else that goes into education. There are many factors that go into a student’s performance: is a test score enough to determine the value of a teacher? Some schools sort students, meaning better teachers often get better students which can widen the test score gap and keep teachers and students from improving. Can the standardized test still be used efficiently as a yardstick for school performance? How reformed can American education be if the improvement simply an illusion based on standardized test scores? How can schools expect honest and solid improvement within the student body when the salaries of their faculty depend only on the results of a single test? The importance seems to lie in the framing and the evaluation weight: test scores might still an integral part of determining teacher value, but placing too much importance on them seems to encourage dishonesty, lack of consistent teaching and too much focus on test scores.