It seems to us they failed: We’re often amazed by the way our biggest newspapers report on the nation’s schools.
So it was this morning, when the New York Times published a 1600-word front-page report about the use of standardized tests to rate New York State teachers.
Governor Cuomo wants to extend the practice. According to reporter Kate Taylor, his proposals, which teacher groups largely oppose, “would both increase the weight of test scores, to 50 percent of a teacher’s rating, and decrease the role of their principals’ observations.”
Should test scores constitute 50 percent of a teacher’s rating? That strikes us as a bad idea. We were struck by Taylor’s failure to state an obvious reason why it seems like a bad idea.
For what it’s worth, we aren’t opposed, as a matter of principle, to the use of test scores in evaluating teachers. We assume that principals have always used test scores in some such way. Consider a hypothetical example from the distant past:
In the spring of 1970, we administered the Iowa Tests of Basic Skills to a class of Baltimore fifth-graders. At that time, the thought didn’t cross our minds that the test results would be used to evaluate us.
That said, suppose our principal noticed that Teacher Smith’s fourth-grade students got horrible test scores year after year. Wouldn’t she have been obligated to figure out why that was happening?
Governor Cuomo wants to go well beyond that. He wants to use student scores in the annual ratings of all teachers.
Near the end of her lengthy report, Taylor presented some objections to this idea. As you can see, her explanation was fuzzy:
TAYLOR (3/23/15): John Bierwirth, the superintendent of the Herricks school district, also on Long Island, where 93 percent of the teachers were rated highly effective, said that in devising his district’s evaluation system, he had intentionally tried to create a cushion to counterbalance the portion of the ratings based on test scores, which for an individual teacher can bounce up and down from year to year.That was pretty much it. Test scores can bounce up and down from year to year! Also, “there are legitimate technical concerns with the value-added scores,” the one quoted expert said.
“I wasn’t gaming the system,” Dr. Bierwirth said, “but I was trying to protect teachers from whimsical results.”
[T]he movement to weigh scores heavily in teacher evaluations has lost some steam. The fact that ratings based on test scores can vary from year to year has led to concern about teachers being unfairly penalized. Additionally, the transition to tougher, Common Core-aligned tests, and the associated drop in scores, has left many teachers, administrators and parents skeptical of the validity of the results. In a Quinnipiac University poll conducted this month, disapproval of the use of test scores helped drag Mr. Cuomo’s approval rating down to 50 percent, his lowest ever.
“Most leaders, even those who support teacher evaluation reform, have decided to reduce the degree to which the evaluations depend on student achievement results,” Michael Petrilli, the president of the Thomas B. Fordham Institute, a conservative education reform organization, said.
“That’s partly to try to make the evaluations more palatable to teachers, but it’s also because they’re trying to make these evaluations more reliable, and there are legitimate technical concerns with the value-added scores,” Mr. Petrilli said, referring to the method by which teachers’ impact on their students’ test results is calculated.
None of those statements are “wrong.” That said, they constitute a very fuzzy tea. Meanwhile, we were struck by the problem which didn’t yodel:
What happens when teachers cheat?
Duh! Unless Cuomo has come up with a very strong security program, that would be an obvious problem with his proposal. Hoping to get a strong evaluation, today’s Teacher Smith might cheat his ascot off with his fourth-grade students.
This means that Teacher Smith will get an inappropriately good evaluation. And the problem doesn’t end there:
The following year, those kids’ test scores will come back to earth when they’re in the fifth grade with Teacher Jones. As a result, Teacher Jones, who didn’t cheat, will get an inappropriately bad evaluation.
Do reporters at the New York Times know that cheating occurs? We’re fairly sure they do! Just last Tuesday, a news report in the Times ran beneath this headline:
“Closing Arguments Begin in Test Cheating Trial of 12 Atlanta Educators”
In the past few years, cheating scandals have been so huge that even our most famous newspapers have managed to report them. But by force of habit and dint of culture, reporters still fail to connect the dots when it comes to a topic like this.
Has Governor Cuomo thought about this? We don’t have the slightest idea! Our mightiest paper, the New York Times, seems disinclined to ask.