The devil's in the details.

It all sounds so simple to most people. Test kids on what they know, then base their teacher's evaluation on what they've learned shown by the test. Those teachers with good "value added" results get tenure and higher pay. Those with poor results get shown the door.

If you believe that the problem with K-12 education in America is infiltration by gangs of bad teachers who don't give a damn whether their students learn or not, this sounds like a solution. If you believe this, you probably haven't spent much time in an American school. The vast majority of teachers desperately want to  see their students learn, but teachers don't control the system. (See Let's stop blaming the teachers!)

But, for the moment, let's talk about "value added" performance evaluation for teachers. Just what standardized test will we use? The New York Times recently ran a story by Michael Winerip pointing out some problems. "For teachers in subject areas and grades that do not have state tests (music, art, technology, kindergarten through third grade) or do not have enough state tests to measure growth (every high school subject), it is the state’s responsibility to create a system of alternative ratings. In New York, that will have to cover 79 percent of all teachers, a total of 175,000 people. The only state tests for assessing teachers are for English and math, from fourth grade to eighth." [Emphasis mine.]

Not wanting to appear "bossy," federal officials have left the nuts and bolts of evaluation up to state officials who are, in general, passing the problem along to local officials. With the exception, that is, of how much the "value added" score should count towards the teacher's total evaluation grade.

"In May 2010, the teachers’ union and department officials, including [NY State Education Commissioner] Dr. King, agreed that student scores on state tests would account for 20 percent of a teacher’s evaluation. In August 2010, Mr. Duncan visited the state union’s headquarters in his Race to the Top bus (he really has one) and told union and department officials that New York had won a grant “because of your collective leadership, your act of courage.” [Emphasis mine.]

"In May 2011, with no warning, Dr. King and Gov. Andrew M. Cuomo rammed a measure through the Board of Regents making state tests worth up to 40 percent of teacher evaluations. In August, a state judge ruled that they couldn’t do that." [Emphasis mine.]

"For the last month now, as federal officials have pressed for a resolution, the governor and the commissioner have been berating the union. Like children who change the rules in the middle of the game, they appear to be counting on a lot of screaming to distract the crowd." [Emphasis mine.]

Back to the question of how to come up with a numerical score for the "value added" performance of teachers in fields such as music, art, physical education, special education, technology, science, social studies, and high school subjects. Do we create paper and pencil standardized tests? That's very expensive. What kind of written test would we give for second grade gym?

Apparently, NY has sent out guidelines. Let's take band as an example. "Several weeks ago the state sent out a guide. The band teacher could listen to every child play at the start of the year and assign a score from 1 to 4. At the end of the year,” the state guide says, “the teacher re-evaluates their students.” (Someone needs to evaluate the state’s grammar.) The teacher again grades students from 1 to 4, and the sum of the progress they have made during the year determines the teacher’s rating." 

Paul R. Infante, the director of fine and applied arts for the Commack School District on Long Island sees some problems with this system. "There is such a variety of ability, he said, that setting a fair baseline at the start of the year would mean assigning children a wide range of music pieces to perform. Just to find the appropriate pieces, he said, the band teacher would have to listen to each child play. A child could be terrible at sight reading but have a nice sound. So in fairness, the teacher would have to spend a few weeks helping 100 children prepare pieces just so they could be tested for their initial rating.
'It would take so much time away from instruction to focus on the assessment,' Mr. Infante said."

Mr. Infante goes on to suggest that perhaps outside evaluators would need to be hired to "...assess the accuracy of the ratings a teacher gave, to make sure they were not artificially low at the start of the year or artificially high at the end." All this while schools are laying off staff.

The state, of course, will see that things go smoothly. The catch here is that "the State Education Department’s budget had been reduced 40 percent in the past few years, staffing [is] thin and the ultimate responsibility for monitoring [will] be left to principals, superintendents and school boards. The main state role, he said, will be to 'provide guidance and models.' "

Well, what could possibly go wrong with this plan?