Corrected: An earlier version of this story misstated the name of the National Council on Teacher Quality.
Bolstered by new research and federal incentives, experts decided about a decade ago that better teacher evaluation was the path to better student achievement. A flood of states started toughening their teacher-evaluation systems, and many of them did it by incorporating student-test scores into educators鈥 ratings.
And while those policies are still in place in a majority of states, there are signs the tide is turning: Over the past two years, a handful of states have begun reversing mandates on using student-growth measures鈥攁nd standardized-test scores, in particular鈥攖o gauge teacher quality.
Six states鈥擜laska, Arkansas, Kansas, Kentucky, North Carolina, and Oklahoma鈥攈ave now dropped requirements that evaluations include student-growth measures and begun letting districts decide what elements to include in assessing teachers, according to analyses from the Education Commission of the States and the National Council on Teacher Quality.
Connecticut, Nevada, and Utah passed policies that require some evidence of student learning but prohibit using state standardized-test scores for that purpose. Florida kept student-growth measures but now lets districts choose how they鈥檙e calculated.
Those are all 鈥渟ignals [states] are backing away from the inclusion of student-growth or value-added measures,鈥 said Stephanie Aragon, a policy analyst for ECS.
The changes are due, at least partly, to the increased flexibility that states have under the new federal education law, the Every Student Succeeds Act.
Some analysts, though, say it鈥檚 still unclear whether states will follow any sort of trend on teacher evaluation.
鈥淚t is playing out and will play out differently across the 50 states,鈥 said Patrick McGuinn, an associate professor of political science and education at Drew University in New Jersey. 鈥淭he forces at play here are pushing in a couple different directions around teacher evaluation.鈥
鈥楾ook the Hook Too Quickly鈥
Teachers鈥 unions have generally fought the use of test scores in teacher evaluations, particularly when those evaluations lead to decisions about teachers鈥 tenure, pay, and dismissal.
The American Federation of Teachers and the National Education Association filed more than a dozen lawsuits nationwide between 2011 and 2015 related to teacher-evaluation systems. The AFT briefly championed the slogan 鈥淰AM is a sham,鈥 referring to value-added measures, which use complicated algorithms to determine how much a teacher contributed to students鈥 academic growth over a year.
Arguments against using student-growth measures, as well as hang-ups encountered in implementing such systems, have been persuasive in some states.
鈥淓ven for those who ultimately believe in the value of measuring outcomes and even accountability systems somehow linked to outcomes, the eagerness to jump really quickly on value-added measures and test-based accountability was premature,鈥 said Jeffrey Henig, a professor of political science and education at Teachers College, Columbia University. Among some state leaders there is 鈥渁 recognition 鈥 that we took the hook too quickly and too eagerly here.鈥
At the same time, legislatures and state school boards that pushed for the inclusion of test scores may be reluctant to turn their backs on those policies too soon. In some places, the evaluation systems were just getting going when ESSA was passed in December 2015.
鈥淚 don鈥檛 think anyone [is] overly enthusiastic to undo something they鈥檇 worked four to five years to roll out,鈥 said Michelle Exstrom, a program director at the National Conference of State Legislatures.
In weighing whether to change these evaluation systems, state leaders are likely considering the time, energy, and resources they鈥檝e put into developing them. 鈥淭here鈥檚 a lot of sunk costs there,鈥 said McGuinn. 鈥淎nd that dynamic works to sustain these systems for a bit.鈥
Annual teacher evaluations were traditionally based on information from a single source: observations from principals.
Starting in 2009, a confluence of factors led to more than two dozen states stiffening their teacher-evaluation requirements. That year, TNTP (formerly the New Teacher Project) published a seminal report called 鈥The Widget Effect,鈥 which found that 99 percent of all teachers were being rated as 鈥渟atisfactory.鈥 Policymakers and education leaders began questioning the validity of evaluation systems that failed to distinguish among teachers.
The Obama administration began its Race to the Top program toward the end of that year. The competitive-grant program offered states financial incentives to include student-test data in their evaluation systems.
At about the same time, the Bill & Melinda Gates Foundation began pouring millions of dollars into studying teacher quality. (91直播 currently receives financial support from the Gates Foundation for coverage of continuous improvement strategies in education.) The foundation鈥檚 high-profile 鈥淢easures of Effective Teaching鈥 study was among the largest randomized experiments of its kind, collecting data on 3,000 teachers across six large districts in order to compare different methods for gauging teacher performance.
And then there were the waivers. Starting in 2011, the U.S. Department of Education began offering states relief from some of the stringent requirements in what was then the main federal education law, the No Child Left Behind Act. (Among other provisions, the law mandated that all students perform at grade level in reading and math by 2014.) To get that flexibility鈥攚hich most states ultimately did鈥攕tates had to commit to linking student-achievement outcomes to their teacher-evaluation systems.
With all those incentives in place, the number of states using student-growth data in their evaluations skyrocketed, going from just 15 states in 2009 to 43 at the end of 2015, according to NCTQ.
Despite making those commitments, states actually implemented the policies at different speeds. Tennessee, for instance, was an early implementer of this kind of evaluation system鈥攁nd hit roadblocks right away. Nevada, on the other hand, passed a law in 2011 requiring the use of student-test scores but had yet to start incorporating the data five years later.
ESSA Offers Reprieve
But with the 2015 passage of ESSA, states almost immediately got a reprieve.
The bipartisan law put teacher evaluation back in states鈥 hands鈥攊n essence renouncing the Obama administration鈥檚 push for strict test-based accountability.
While six states dropped requirements around using student-growth in evaluations, they did so in different ways.
Some, like Arkansas and Kentucky, did so through state legislation. In fact, the National Conference of State Legislators has tracked bills in 10 states that proposed such changes.
鈥淲e are just now starting to see the effects of ESSA on state legislation, and we don鈥檛 anticipate seeing the majority of it till next legislative session,鈥 said Exstrom of NCSL.
A few other states, including Alaska, Connecticut, and North Carolina, passed policies backing away from student-growth requirements in teacher evaluations through their state boards of education.
Before ESSA, Kentucky required that student growth be a significant factor in teacher evaluations鈥攁nd didn鈥檛 allow teachers to receive a high final rating without a high rating on student scores. In 2017, it passed a bill that allowed districts to decide whether to include student achievement measures at all.
So far, though, 鈥渁 lot of districts have chosen not to make huge changes to their systems,鈥 said Robin Hebert, the director of the division of next-generation professionals at the Kentucky education department. That鈥檚 likely because they鈥檙e waiting for the specific state regulations to be released, which should happen in the spring.
The Kentucky Education Association, the state NEA affiliate, for its part, is pleased with the change.
鈥淜EA has always felt that student growth should not be one of the multiple measures included in the teacher-evaluation system,鈥 the group鈥檚 president Stephanie Winkler wrote in an email.
However, Elizabeth Ross, the managing director for state policy at NCTQ, which advocates for measuring teacher effectiveness through objective data like test scores, called the Kentucky change 鈥渁 huge step backward for them.鈥
In Arizona, Maine, and New Mexico, the legislatures approved bills to undo or reduce the weight of the student-achievement requirement鈥攐nly to have them vetoed by their governors.
New Mexico, well known to have the toughest evaluation system in the country, has been at the center of the debate. The state education chief recently reduced the student-growth measure from 50 percent of a teacher鈥檚 evaluation to 35 percent.
Several other states, including Indiana and Louisiana, have convened task forces to look at the issue.
Changes Ahead?
There鈥檚 another way to view the state policy changes so far, some analysts say: There could have been more.
鈥淪ome folks thought after ESSA, states would rush to undo this,鈥 said Exstrom.
In another sense, the swing toward including student achievement in teacher evaluations really didn鈥檛 change much on the ground.
Just as before, nearly all teachers continued to get positive ratings, even in states that overhauled their evaluation systems, a recent study showed. (New Mexico, where nearly 1 in 4 teachers were rated 鈥渋neffective,鈥 is an outlier.) That鈥檚 largely because principal observations still make up the bulk of the evaluations nationwide鈥攁nd principals almost never give bad reviews.
鈥淭here was some thought that if you [toughen evaluation systems], a whole bunch of teachers out there will be able to demonstrate they aren鈥檛 able to do their job,鈥 said Exstrom. 鈥淚 think that just hasn鈥檛 come to light. ... It hasn鈥檛 caused the big traumatic effects some thought would happen.鈥 In fact, many states are more concerned with teacher shortages than they are with evaluation policies, she said.
Even so, the fate of teacher evaluations in many states may be dependent on the 2018 election results.
鈥淧robably the more Democratic victories you see in state legislatures and governorships, the more likely you are to see teacher-evaluation reforms rolled back to one extent or another,鈥 said McGuinn. 鈥淲hat the electoral results are in states and at the national level in 2018鈥攃ertainly, it matters.鈥