Many teachers and their unions have taken strong public positions against any use of student test scores in teacher evaluations. Until recently, however, many were open to the idea, which raises some critical questions:
- Can teacher quality be effectively evaluated without some objective measure of student success?
- If teachers aren’t accountable for boosting student learning as measured by valid assessments, what are they accountable for?
- If test scores and student growth are ruled out of teacher evaluations, what factors should be considered to get the most effective teachers in front of the kids who need them most?
- Finally, will resistance to test-based accountability undermine their case for more resources and their desire for more autonomy and respect?
In her excellent book, The Teacher Wars, Dana Goldstein recounts how in the 1980s, both national teachers unions responded to calls for merit pay by insisting that it be based on valid “objective” measurements. Goldstein quotes teachers union leader Albert Shanker bluntly rejecting a system where, “Some principal walks into your room and gives a judgment as to who is a better teacher.”
When the Obama administration took office, everyone from teachers unions to Arne Duncan believed that teacher evaluation in the U.S. was “broken” because it failed to provide useful feedback on instructional practice, or identify and reward excellent teaching. In most school districts, close to 100 percent of teachers were rated “satisfactory,” making teacher evaluation a national sham.
The Obama Administration’s multiple measures policy calls for student growth on standardized tests to be one of several factors in a teacher’s evaluation—along with observation, student work, etc. Initially, many teachers unions were on board.
In fact, 65 percent of the local teachers union affiliates in the 10 states that won second-round Race to the Top grants signed applications supporting the evaluation policy. Hundreds of additional affiliates from states that did not win signed similar letters of support.
The first two winners, Delaware and Tennessee, had the support of 100 percent and 93 percent of their local unions respectively. Lillian Lowery, Delaware’s state education chief at the time and current Maryland state chief said, “Our teachers unions were at the table every step of the way…Compromise is not a bad thing.”
The website of the Tennessee Education Association describes how a committee of teachers helped craft the state’s evaluation policy, under which, “50 percent of the new evaluation must consist of student achievement data, of which 35 percent will be Tennessee Value Added Assessment System data or some other comparable measure of student growth.”
National union leaders were also supportive of teacher evaluations that included student growth on tests. In a January 2010 speech at the National Press Club, AFT President Randi Weingarten said that teacher evaluations should be “based on multiple measures, including student test scores based on valid and reliable assessments that show students’ real growth while in the teacher’s classroom.”
In 2012, Weingarten and National Education Association President Dennis Van Roekel signed a “vision statement” explicitly supporting evaluations based on “measures of academic student growth.” Their co-signers included U.S. Secretary of Education Arne Duncan and the leaders of organizations representing school boards, superintendents, large urban districts, state chiefs and federal mediators.
Even the Chicago Teachers Union, flush off of its first strike in 25 years, signed a 2012 contract that included test scores in teacher evaluations. According to the contract, “Student growth scores shall be 25 percent of the summative rating in school year 2012-2013, 25 percent of the summative rating in school year 2013-2014 and 30 percent of the summative rating in school year 2014-2015.” Other districts like New Haven in Connecticut, Hillsborough County in Florida, and Washington, D.C., also embraced evaluation of teachers based in part on measures of student growth.
Before the policy has even had much chance to work, states are delaying implementation and the policy as well as the tests are under siege. President Weingarten frequently criticizes value-added metrics as unreliable. NEA President Lily Eskelsen-Garcia decries “toxic testing” at every turn. Her union has launched an ad campaign against standardized testing.
The New Jersey teachers union has also launched an opt-out campaign, as has the statewide teachers union in New York. New York State United Teachers President Karen Magee openly admits that her goal is to undermine evaluation.
Unions are certainly entitled to change their minds on an issue as complex as evaluating teachers, but let’s at least acknowledge that many of them previously supported multiple measures. Critics of the policy also have an obligation to come up with a practical, fair and objective alternative that both strengthens the teaching profession and serves the interests of children. So far, they haven’t.