ON MAKING A RESEARCH-BASED TEACHER EVALUATION SYSTEM
New Studies Show the Value of Great Teachers and the Best Ways to Identify Them
While Senator Ruiz (D-29) has been drafting her education reform bill, two very credible, important studies have delivered some highly relevant findings drawn from massive amounts of data. We hope this groundbreaking research on teachers and evaluations gives NJ legislators the courage to do what’s right for NJ kids.
Before delving into the studies, themselves, it is important to point out that one of the major claims of reform opponents is that there is no research to back up reform proposals. These two recent, exhaustive studies show that claim to be demonstrably false.
The Research Studies
The first study is by Harvard/Columbia economists of 2.5 million students over a 20-year period and is discussed in Eric Lerum’s guest blog last week (here). After analyzing a truly staggering amount of data, the researchers conclude that teacher effectiveness can be measured by using “value-added” analysis of student achievement growth on standardized tests.
Moreover, effective teachers not only have significant and lasting impacts on their students’ performance in school, they also have significant impacts on these students’ life outcomes. Students assigned to high-value-added teachers were more likely to attend college, earn higher salaries, live in better neighborhoods, and save more for retirement; and were less likely to have children as teenagers. President Obama used the results of this study in his State of the Union address last night.
At the same time, the Gates Foundation released its Measures of Effective Teaching (MET) report (here). This two-year project of “unprecedented scope” involved 3,000 teachers in six districts across the country. The New Teacher Project (TNTP) wrote a companion guide, “MET Made Simple,” that outlined the most important findings and policy recommendations.
The four “key” lessons in the MET report were:
· Value-added analysis is more accurate than any other single measure in predicting a teacher’s long-term contributions to student success. Teachers with high value-added scores are not “teaching to the test” but actually improve students’ higher-level thinking, and these students report trying harder and enjoying school more
· Classroom observations give teachers valuable feedback on how to improve but are of limited value in predicting future performance at helping students learn. More observations and observers are better.
· Most teachers are able to manage their classrooms but lack basic instructional skills, such as teaching techniques or communicating with students. There appears to be a wide gap between evaluation ratings and actual classroom performance.
· Evaluations that combine several strong performance measures will produce the most accurate results. Such multiple measures should include: value-added data, classroom observations and student surveys.
As the NJ legislature considers education reform, the following recommendations emerge from these two landmark studies:
· Base teacher evaluations on multiple measures of performance including “value-added” data on student academic progress. The studies show that a high value-added score is the best indicator of a great teacher, and that great teachers not only benefit their students in school, they benefit their life outcomes.
· In neither study did high value-added teachers “teach to the test.” In grades and subjects where common assessments are not available, states and districts should develop them.
· Observations alone are insufficient. They give important feedback but are not good predictors of future teaching success. The more the number of observations and observers, the better.
· Give teachers and evaluators the training and ongoing support they need to be successful. Observations are only as good as the observers. Teachers need continuous professional development to ensure that they have the instructional skills they need.
· Consider using student surveys as part of teacher evaluations. Students know effective teaching when they see it and student surveys were as correlated with student learning as teacher observations. Such surveys are especially useful where there are no standardized tests.
The research is there and the legislation has been introduced. The time is right for NJ lawmakers to pass meaningful and beneficial education reform.