PAA Policy Brief: End Test-based Teacher Evaluation
Parents call for states to take back control of teacher evaluation,
end misuse of test scores
The Every Student Succeeds Act, which became law in December, has given states new control over education policy. Parents across the nation call on state policy-makers to use their restored authority to rein in the high-stakes standardized testing that plagues American education.
Washington still requires annual reading and math tests in grades 3-8, science tests in grades 5 and 8, and three tests in high school. But state leaders now have the power to eliminate dozens of other tests – particularly those whose main purpose is rating teachers. Policy-makers should overhaul teacher evaluation systems immediately.
Starting in 2009, the U.S. Department of Education began pressuring states to add a test-based measurement to teacher effectiveness ratings. There was no solid evidence that the requirement would improve educational opportunities or outcomes. Still, many states bowed to federal pressure. They began to give teachers “growth” or “value added” (VAM) ratings calculated from student standardized test scores.
Producing a VAM rating for every teacher requires a standardized evaluation for every course. As a result, states and districts have spent hundreds of millions of dollars to develop, purchase, administer and grade additional tests. Students and teachers have spent countless hours preparing for and taking those tests.
These efforts have not produced improvements that justify the time, money and energy invested in them. Nor are they likely to, for several reasons.
• The standardized test results used to calculate most VAM scores involve only a few of the skills that twenty-first century students need. According to the National Research Council, standardized tests cover a limited range of knowledge and abilities. They generally omit “the portion of the curriculum that deals with higher levels of cognitive functioning and application of knowledge and skills.”
• Studies of teacher effects on test score gains or losses indicate that only 10-15 percent of the variation in student academic growth is due to variation among teachers.
• VAM calculations are plagued by widely documented technical problems. The number of student test results used to calculate individual teachers’ value-added ratings is generally too small to be statistically meaningful. Even seemingly simple tasks such as assigning student scores to the correct teacher can be quite difficult. Not surprisingly, studies have shown that teachers’ value-added ratings often swing dramatically from year to year.
• The proliferation of high-stakes standardized tests has had many negative side effects. Over-testing has led to increased stress, a narrowed curriculum, and widespread teaching to the test. It has caused students to lose interest in school and learning, driven excellent teachers from the profession, and discouraged young people from pursuing teaching careers. It has fueled the school-to-prison pipeline, sparked cheating scandals, and diverted time, energy and resources from other educational goals. These negative effects have been especially evident in schools that serve low-income children of color. Even if VAM ratings could be calculated accurately, the limited information they provide is simply not worth the effort or the consequences.
In large-scale scientific studies, such as tests of new drugs, periodic review ensures that experimental treatments are not harming participants. If it becomes evident that a new treatment is doing more harm than good, the study immediately ends. In the case of VAM, the evidence is clear. State leaders should act boldly, and stop the damage now.
References for the research in this policy brief, and a downloadable pdf version of the full paper with references, are here.
A one-page fact sheet on the dangers of value-added measurement (VAM) in teacher evaluation is here.