When the Every Student Succeeds Act removed the federal requirements from teacher-evaluation systems imposed by the Obama administration, states were given the opportunity to decide the best way to grade their teachers.
And a year after ESSA鈥檚 passage, most states are still tinkering with their evaluation systems鈥攑articularly regarding student growth measures, a controversial addition that assesses teachers based on students鈥� academic progress and performance.
ESSA does not require states to set up teacher-evaluation systems based in significant part on students鈥� test scores, which was a key part of the U.S. Department of Education鈥檚 state-waiver system under the No Child Left Behind Act, the predecessor to ESSA.
States now have a newfound flexibility to adjust their evaluation systems鈥攁nd in doing so, they鈥檙e all over the map.
鈥淚f I were to use two words to describe the landscape right now in regard to teacher accountability: messy and complicated,鈥� said Kimberly Kappler Hewitt, an assistant professor of educational leadership at the University of North Carolina at Greensboro, who researches teacher-evaluation policies and student-growth models. 鈥淣o two states are doing the same thing right now.鈥�
A number of states, however, have been moving away from student growth measures in evaluations, Kappler Hewitt said, whether that means postponing their inclusion, reducing their percentage in the evaluation breakdown, or eliminating those measures altogether.
Not all of that is directly related to ESSA, said Andy Baxter, the vice president of educator effectiveness at the Southern Regional Education Board. States had been moving away from student growth measures for the past several years. 鈥淭he ESSA requirement [change] will only accelerate that or reinforce that,鈥� he said.
Still, some states, including Iowa and Alabama, have at least considered tying student test scores to evaluations.
鈥淚 don鈥檛 think we鈥檙e going to have any sort of consensus or any sort of mellowing out of the waters for several years as states try to grapple and feel their way through these things,鈥� Kappler Hewitt said.
The Policy Landscape
Even within states, there has been some back and forth.
In 2012, the Alaska State Board of Education had approved a plan that would eventually require 50 percent of a teacher鈥檚 evaluation to be comprised of student test scores. But the board delayed the implementation and ultimately repealed the requirement in June.
In August, the New Jersey Department of Education tripled the weight of PARCC tests in teacher evaluations鈥攖o 30 percent. The decision drew criticism from state teachers鈥� unions and some local districts, and legislators quickly introduced a bill that would eliminate student standardized test scores from teacher evaluations.
The bill passed the New Jersey General Assembly and is currently pending in the state Senate鈥檚 education committee.
In another policy turnaround, New York in December 2015 suspended the use of student scores on statewide tests in teacher evaluations for four years鈥攍ess than a year after the governor signed a law that increased the weight of test scores to up to 50 percent in evaluations. Teachers in New York will still receive a growth score based on tests, but it won鈥檛 count toward employment consequences, positive or negative. By 2019, state officials have said, the teacher-evaluation system will be completely revamped.
Kappler Hewitt said New York exemplifies one of the trends she is seeing: putting the brakes on test-score data for the time being.
鈥淪tates have gone through a lot of change in terms of standards and assessments,鈥� she said. 鈥淎ll that change, that churn, has big implications for teacher evaluations.鈥�
For example, Ohio will not use value-added ratings from state tests for the 2014-15 and 2015-16 school years in evaluations, to ease the transition to new state tests.
Meanwhile, Alabama considered moving toward incorporating student growth measures into teacher evaluations with a controversial piece of legislation that died in committee in last year鈥檚 session. The bill would have tied 25 percent of an evaluation score to student growth, using the ACT and the ACT Aspire college-readiness tests as indicators of student performance.
The bill was largely condemned by educators in the state, including Jennifer Brown, the 2015-16 Alabama teacher of the year, who led a social media campaign against it.
鈥淜ids are more than a score,鈥� she said. 鈥淎 test score doesn鈥檛 determine how well I taught that student. I just don鈥檛 think that鈥檚 a good measure about how well your student is doing.鈥�
Brown was also concerned about the logistics. How would educators who don鈥檛 teach core subjects, for example, art or physical education, be evaluated? Furthermore, the ACT and the ACT Aspire don鈥檛 count for a student鈥檚 grade. 鈥淚鈥檓 very much opposed to using a test for teacher accountability when there鈥檚 no accountability for the student,鈥� she said. 鈥淚t鈥檚 got to be fair for educators. Until there鈥檚 a fair system, I鈥檓 going to oppose using tests for evaluation.鈥�
In late September, California Gov. Jerry Brown vetoed a bill that would have required school districts to evaluate teachers based on multiple measures of student academic growth and performance, including state standardized test scores.
And in the spring, North Carolina removed student growth measures as a stand-alone standard in the state鈥檚 teacher evaluation system. 鈥淲e wanted growth as a school improvement tool, and not as a hammer,鈥� said Rebecca Garland, the deputy state superintendent for the state鈥檚 department of public instruction.
The data, which are calculated from student assessment scores, were never meant to be punitive, she said. They鈥檙e meant to assist principals in determining professional development and to counsel teachers about how they can improve.
But 鈥渢eachers were becoming overly anxious about whether they would get fired or not,鈥� said Thomas Tomberlin, the director of district human resources at the state education department. "[We heard] feedback from the field that teachers were not feeling like they could focus on the improvement aspect of the data because they were too concerned about possible employment ramifications.鈥�
Now, teachers and principals will still receive the growth data鈥攂ut how it is integrated into the teacher evaluation system is at the discretion of individual districts, Tomberlin said.
鈥淚 think teachers, by and large, have come to see the value of the information鈥攖he limited value, it doesn鈥檛 tell us everything鈥攁nd teachers are much more comfortable with it as a data point than they have been in the past,鈥� he said.
Setting Up Safeguards for Equity
Despite the tinkering within states, Daniel Weisberg, the CEO of the teacher-training and advocacy group TNTP, said he doesn鈥檛 see states backing away entirely from considering student growth measures under ESSA鈥檚 new flexibility. In the No Child Left Behind era, some states already had committed to retooling their evaluation systems to qualify for federal Race to the Top funds.
The goals of that approach鈥攖o provide teachers regular feedback on their performance against clear performance standards鈥攁re still widely supported by many state officials and others who want hard and fast accountability based on metrics, Weisberg said.
鈥淵ou haven鈥檛 seen the clock turn back to 2009, and I don鈥檛 think you鈥檙e going to see that,鈥� he said.
Not all educators are convinced that student growth measures are the way to ensure equity, but SREB鈥檚 Baxter said it is a challenge for states to make sure that groups of students are not systematically getting teachers with weaker skills.
鈥淚 think states do need to hold districts, and in turn their schools, accountable for some sort of connection between evaluation ratings of the teachers and the success of students,鈥� Baxter said. 鈥淗aving student growth is one way to do that, having student surveys is another way. In the absence of those, we鈥檝e got to avoid situations where large numbers of teachers in the school are getting high ratings and students aren鈥檛 growing. ... States need to set up safeguards.鈥�
North Carolina鈥檚 policy of providing teachers and principals student growth data while not mandating its inclusion in evaluations is one such safeguard, he said. Another is Louisiana鈥檚 public database of teachers鈥� evaluation ratings by school. If there are schools with a large percentage of highly effective teachers, but students aren鈥檛 progressing, that鈥檚 a problem, Baxter said.
Weisberg said he hopes states will continue to 鈥減rovide teachers and principals and school system leaders with meaningful information on how teachers are doing and growing and where they need help.鈥�
But, he said, he doesn鈥檛 expect there to be a consensus among states, when the dust settles: 鈥淭here isn鈥檛 a magic bullet answer.鈥�
