California Says No to Teacher Evaluation Reform

Huck/Konopacki Labor Cartoons

As the Obama Administration gets ready to offer waivers to states for No Child Left Behind, it is looking more and more like states will be required to accept a number of onerous conditions to win these waivers, including tying teacher evaluations to student test scores. This is Obama’s stick to beat teachers into submission. A few years ago he offered a carrot that looked very similar: use test scores to evaluate teachers and ease restrictions on charter schools in order to be eligible for some of the $4.3 billion in federal Race to the Top grants.

There goes crazy California again.

California is poised to buck this trend and many conservatives are rolling their eyes at “crazy California,” (which just got crazier with the passage of the rectal injection law).

AB5, which would have required school districts to use student test scores as part of teacher evaluations, died in the state assembly last week, according to the Orange County Register. However, John Fensterwald - Educated Guess, says that in reality the vote on the bill was merely postponed for a year.

Regardless, it is not California, but the test mongers who are crazy for believing that student test scores are an accurate measure of teacher quality.

They are crazy for several reasons. First, student test scores are a measure of students’ ability to pass a multiple choice standardized exam, not a measure of teachers’ skill in the classroom. Students may perform poorly on these exams for numerous reasons that have nothing to do with teachers like test anxiety, insomnia, stress at home, or apathy toward the test itself. Often the tests have no bearing on graduation or college entrance, yet take up entire school days or weeks, making students feel burned out, oppressed and disinclined to take them seriously. However, even when students take the tests seriously their scores are correlated more strongly with their socioeconomic backgrounds than any other factor, including their schools and teachers. Affluent students tend to get higher scores and show stronger improvement when schools make changes to improve test scores.

Perhaps the craziest aspect of this plan is that states want the NCLB waivers precisely because the tests are such a terrible measure of anything. Secretary of Education Arne Duncan says that as many as 80% of all American schools will soon be failing under NCLB’s interpretation of these test scores, while the Orange County Register says that California state superintendent of schools Tom Torlakson expects a similar number of failures in California. Torlakson, to his credit, asked Duncan for an unconditional waiver. His reasons, however, had nothing to do with the irrationality of the tests and their assessments. Rather, he argued that it would be unfair for the federal government to demand overhauls without providing money to implement them.

The high failure rate is not just due to increasing rates of childhood poverty, which is certainly an important factor, but also to the very design of the assessment. NCLB deems a school a failure if any one subgroup of students does not make “adequate yearly progress (AYP),” even if all other subgroups do. For example, if boys, girls, special education, English language learners all improve, but those on free or reduced lunch do not, the entire school is considered failing. Yet if the majority of subgroups are improving, it would suggest that the quality of teaching is good and the schools’ interventions are, for the most part, working.

Another problem with the testing scheme is that the scores are a moving target. If a school improves over the previous year, but does not improve as much as other schools, it will not meet its AYP and could still be deemed a failure. Considering that the law requires students to be “proficient” in math and English, it would be much more appropriate, honest and meaningful to set fixed benchmarks that students must meet rather than comparing them with each other.

Who’s Running the Asylum?

Despite the insanity of tying teacher evaluations to student test scores, over two dozen states now use students' scores on standardized tests to evaluate teachers and many even tie pay raises, promotions and the loss of tenure to these scores. In Indiana, Ohio and Michigan, for example, school districts can promote and fire teachers based on students’ performance on state tests and in Washington state teachers are financially rewarded for student academic gains.

Et Tu NEA?

The teachers unions have largely opposed tying teacher evaluations to NCLB tests. However, in July, the nation's largest teachers union, the 3.2 million-strong National Education Association (NEA), said it would support test-based evaluations if the tests were “developmentally appropriate, scientifically valid, and reliable for the purpose of measuring both student learning and a teacher's performance.” This was a terrible concession as there is no such thing as a student test that measures teacher performance. If we want to know how good teachers are then we need to assess them, not their students. If we depend on an unreliable proxy like existing state tests, many good teachers who happen to teach in low income schools will be mischaracterized as bad teachers, while some mediocre and bad teachers who happen to teach in affluent schools will fall through the cracks.

The NEA’s move was also a capitulation to the anti-teacher hysteria that dominates the public education discourse. At the very root of the teacher evaluation debate is the assumption that there are too many terrible teachers that must be punished or fired. Yet there is no valid evidence that this is the case. Furthermore, the hysteria is being promoted by politicians and capitalists who want to weaken or crush the teachers unions and the labor movement as a whole. Rather than accepting the assumption that there are a lot of bad teachers requiring discipline or firings, the unions should be challenging the assumptions of the Ed Deformers and attacking their irrational metrics.

Worst of all, by even engaging the education deformers in this game, the teachers unions are complicit in the continued obfuscation of the biggest cause of the achievement gap—the wealth gap. The teachers unions ought to be investing the vast majority of their resources and energy into public outreach and organizing around this issue. In fact, ending poverty will result in the largest educational gains that can possibly be achieved, much more so than improving teachers or schools.