At a time when teachers鈥� unions are pushing back on newly redesigned evaluation systems, the debates are starting to find their way into the courts.
The latest action comes in Tennessee, where the state union filed a federal lawsuit March 21 calling the portion of the state鈥檚 teacher-evaluation system based on test scores arbitrary, flawed, and in violation of teachers鈥� constitutional rights. Last year, a Florida union also sued over that state鈥檚 system.
The lawsuit this month is the second in the Volunteer State to challenge performance bonuses based on such calculations, and it may signal a rash of challenges nationwide.
鈥淭he legal terrain is not settled, and it has to be settled in almost every state separately because so much of this is driven by state law,鈥� said Douglas N. Harris, an associate professor of economics at Tulane University in New Orleans, who has studied 鈥渧alue added鈥� calculations like those used in Tennessee. 鈥淭hat鈥檚 one reason I think this is only the beginning.鈥�
See related story, 鈥淭ennessee Board Rescinds Plan to Tie Licenses to Test Data.鈥�
Both the state鈥檚 Republican governor, Bill Haslam, and its education commissioner, Kevin Huffman, are named in the suit, as is the Knox County board of education. The case was filed in the U.S. District Court for the Eastern District of Tennessee, in Knoxville.
Spurred by a bonus that a Knox County teacher didn鈥檛 receive as a result of the test-score calculations, the suit has ambitious aims. Among other objectives, it asks the court to prevent the state from using test scores in any employment decision--whether evaluations, pay, or dismissal--until the 鈥渁rbitrary classifications鈥� in the system are corrected.
鈥淧otentially, teachers could lose not only the bonus, but also other forms of compensation, perhaps lose their jobs, certainly lose their eligibility for tenure,鈥� said Gera Summerford, the president of the Tennessee Education Association, an affiliate of the 3 million-member National Education Association.
Officials at the state education department would not comment on pending litigation. In past appearances, though, Commissioner Huffman has attributed the state鈥檚 recent gains on the National Assessment of Educational Progress in part to improved instruction spurred by the evaluation system.
Different Classifications?
In general, value-added models use statistical formulas to generate estimates of how much a particular school or classroom teacher contributed to student learning, as measured by standardized-test scores. Tennessee鈥檚 version is known as TVAAS. It is included as one factor in a teacher-evaluation system that was rolled out, somewhat bumpily, in the 2011-12 school year.
Under state rules, a teacher with five or fewer tested students is graded on a 鈥渟choolwide鈥� measure, based on the progress of all students in the school. Those teachers with six or more tested pupils receive an individual value-added estimate based on those students鈥� progress.
In the case of plaintiff Mark Taylor, an 8th grade science teacher, the value-added score was based on just the 22 students in his regular science class. They represented fewer than 16 percent of the total number of students he instructs, because he also has four sections of students who take an advanced course that does not conclude with a standardized exam.
The value-added system gave Mr. Taylor the lowest possible score for that part of the evaluation. As a result, 鈥渢he plaintiff was denied a bonus under [the district pay program] even though the observation component of his evaluation ... showed that he was exceeding expectations,鈥� the complaint reads.
The union says that the state has 鈥渘o rational basis鈥� for basing the measurement on only a fraction of a teacher鈥檚 students, and that the 鈥渁rbitrary and irrational鈥� categorization of teachers into groups with different evaluation rules violates teachers鈥� due-process and equal-protection rights under the U.S. Constitution.
In an interesting wrinkle, the lawsuit cites as evidence comments made by the developer of TVAAS, the North Carolina-based researcher William Sanders. Mr. Taylor鈥檚 parents were apparently acquainted with Mr. Sanders through Sunday school classes, and queried him by email whether TVAAS results based on one course were appropriate to use for evaluation purposes.
鈥淔or an overall evaluation of the effectiveness of the teacher to facilitate student academic progress, of course not,鈥� Mr. Sanders replied, according to copies appended to the complaint.
The union鈥檚 aggressive stance against the system marks a turnaround of sorts. In 2010, the TEA endorsed the state鈥檚 bid in the federal Race to the Top competition, which explicitly linked students鈥� test scores to teacher evaluations. The support helped the state win some $500 million in the grant competition, but the union now says it didn鈥檛 predict that the changes would lead to so many difficulties.
鈥淲hen I talk to teachers around the state, the number of things [that] can go wrong with TVAAS is the number of teachers in the state,鈥� Ms. Summerford said. 鈥淚t just seems like all these things keep bubbling up that show us the problems it has that we didn鈥檛 anticipate.鈥�
She expects the TEA will file more lawsuits against the use of TVAAS as other teachers鈥� experiences come to light.
The issue of overtesting has, in the meantime, become a general theme for the union. It has launched a statewide campaign, 鈥淭each the Students, Not the Tests,鈥� and in recent months also has put its weight behind legislation to prevent teacher licenses from being granted or revoked based on TVAAS data. (See story, this page.)
National Echoes
The Tennessee lawsuit has echoes elsewhere, as teachers鈥� unions step up their criticism of the use of test scores in evaluations.
In January, Randi Weingarten, the president of the American Federation of Teachers, repudiated her earlier position that such scores could be one factor in evaluations.
And last year, the NEA鈥檚 Florida affiliate filed a lawsuit focused on the value-added segment of that state鈥檚 evaluation system. By using a schoolwide value-added formula, the state was grading many teachers on the performance of students they didn鈥檛 teach, or in subjects they didn鈥檛 teach, the union argued. State lawmakers made some legislative alterations to the system shortly after.
As for Tennessee, even if the court agrees with the union鈥檚 arguments, new forms of teacher evaluation are probably here to stay, said Mr. Harris, the Tulane professor.
鈥淰alue-added was the spark that started these changes to the evaluation systems, but you don鈥檛 need the spark to keep the fire going,鈥� he said.
Supporters of value-added methods, meanwhile, argue that the estimates, while imperfect, are better than many other gauges of teacher quality. Most traditional proxies, such as completing credentialing requirements or holding a master鈥檚 degree, bear an inconsistent relationship to students鈥� academic progress, they note.
