In her commentary piece last week, public education advocate Wendy Lecker returned to the issue of Governor Dannel Malloy and Lt. Governor Nancy Wyman’s unfair, inappropriate and fundamentally flawed teacher evaluation system. Her article, entitled Teacher evaluation system needs overhaul, first appeared in the Stamford Advocate.
While Wendy Lecker has pounded away about the problems associated with Connecticut’s teacher evaluation system for four years, the good news is that it seems that some of the power-elite are finally listening.
Having helped craft and usher in the absurd and destructive teacher evaluation system, the Connecticut Education Association (CEA) will be holding a press conference later today, January 7, 2016, in which they will apparently stand up for Connecticut educators and take a strong stand against Malloy and Wyman’s anti-teacher and anti-public school, teacher evaluation program.
The problem with the existing teacher evaluation system could not be any clearer. As Wendy Lecker explains – Teacher evaluation system needs overhaul;
With the passage of the new federal law replacing the No Child Left Behind Law (NCLB), Connecticut now has a unique opportunity to rethink its flawed teacher evaluation system.
In response first to the federal Race to the Top grant and then the NCLB waiver mandates, Connecticut developed a teacher and principal evaluation system calling for student standardized test scores to be a part of a teacher and principal’s effectiveness rating.
Under the federal law replacing NCLB, the Every Student Succeeds Act (“ESSA”), the federal government no longer requires states to link student standardized test scores to teacher evaluations.
Connecticut’s Performance Evaluation Advisory Council (“PEAC”), the body that oversees the teacher and principal evaluation system, will next meet in January. Since PEAC last met, the notion that one can rate teacher’s effectiveness based on student standardized test scores has been thoroughly debunked.
As the American Statistical Association maintains, teachers account for only 1-14 percent of the variance in student standardized test scores. Joining the ASA and others, the American Educational Research Association recently declared that it is almost impossible to disentangle this tiny teacher effect on student test scores from other in-school and out-of-school factors. A New Mexico court recently blocked that state’s test-based teacher evaluation system because there is no scientific evidence proving that such a system is valid.
Standardized achievement tests were not designed to be instructionally sensitive, i.e. show what or how well a teacher teaches. They cannot be jury-rigged after the fact to be.
In light of the ESSA, some PEAC members, notably the Connecticut Education Association, now indicate they would advocate decoupling statewide standardized tests from evaluations. Indeed, why continue such a demonstrably invalid practice?
Other members, including Connecticut’s superintendents’ and boards of education associations (CAPSS and CABE), maintain that standardized test scores must still be included because they show “student achievement growth.”
What does that mean?
Learning is a complex process. Even if one focuses only on cognitive skills, different grades teach different content and different skills. Each standardized test measures skills that supposedly correspond to that grade level. Comparing one grade level test to another is comparing apples to oranges.
As I wrote in an earlier column (bit.ly/1sOOxFc), in constructing “growth scales” for standardized tests, statisticians make a fictional assumption that learning in math and reading is linear and can actually be compared from year to year. To make this work, they can only focus on a limited universe of skills that might be subject to such a rough comparison.
Measuring growth through standardized tests is, at best, looking at a tiny fraction of cognitive skills.
When we construct an evaluation system based on that tiny universe of disjointed skills, all the components in that system will be equally narrow. Any observations of and conclusions about teachers will center only on how those teachers are teaching those particular skills.
Why do we want to know so little about a teacher?
I want much more for my son. I want my son’s teachers to help him learn skills, but I also want them to help him apply those skills to other subjects and in life. I want them to help him make sense of the world. I want them to help him ask better questions, so he can become a more critical thinker. I want them to help him be a better member of his school community so he can learn to become a good citizen. I want teachers who can assess my child with tools they developed based on their teaching.
None of these teaching skills can be measured with a test.
However test scores are simple, readily available measures; so policy-makers embrace them, even when they are inappropriate.
Rather than construct an evaluation process based on what is easiest to measure, shouldn’t PEAC start with an examination of the type of skills we want in teachers?
Determining whether a teacher has those skills will require us to rely on the professional judgment of administrators and other teachers who observe a teacher’s practices, the work she assigns, and her students’ work.
The state can provide guidelines but it is time start trusting professional educators again. Teaching and learning are complex human endeavors that will never be properly reduced to numbers.
Connecticut now has the opportunity and moral duty to right the wrong being done to our teachers and students. All eyes will be on the PEAC members to see if they have the courage and wisdom to do so.
For more about Connecticut’s flawed teacher evaluation policies check out the following Wait, What? posts;