Vergara v. California: Dismissing Bad Teachers

As we have discussed, Vergara v. California challenges the operations of the Los Angeles Unified School District (LAUSD), which, its plaintiffs claim, disadvantage poor and minority students.

According to the school district, black students in Los Angeles are 43 percent more likely than white students to be taught by a "grossly ineffective" teacher, i.e., one whose performance is in the bottom five percent of teacher rankings.  Hispanic students are 68 percent more likely than whites to have such teachers.

In the 10 years before Vergara was filed, a total of 91 teachers were fired in the entire state of California, most for "egregious" conduct.  Poor performance was a factor in only 19 of the firings.  California public schools employ about 300,000 teachers.

This averages to two people out of 300,000 let go for poor performance every year.

There are two possibilities here:  Either the state had the most phenomenally excellent teaching force in world history, or there was virtually no effort to identify poor teachers and get them out of classrooms.

I vote for No. 2.

The question in Vergara is whether state laws caused this to happen.  The Vergara plaintiffs say California laws have become teacher-friendly at the expense of students and that the processes and expenses of getting rid of poor teachers are so arduous that school districts didn't bother to make the effort.  The defense -- the state and the state's two teacher unions -- claim the problem was school district mismanagement.  They also say schools need more money.

John Deasy

The first witness called in the trial was LAUSD Superintendent John
Deasy, who took his job in 2011.

In the 2006-2007 school year, the LAUSD dismissed three teachers.  The district employed, and continues to employ, almost 30,000 teachers.

In the 2009-2010 school year, LAUSD initiated dismissal hearings for 10 teachers whom the district deemed grossly ineffective. Obviously, given the above numbers, most of those teachers were retained or resigned without taking their cases to court.

In Deasy's first full year on the job, 2011- 1012, the district recommended 99 teachers for dismissal hearings.  Another 122 teachers resigned after being notified they were being recommended for termination and before hearings were convened to examine their cases.

"Just because these 122 resigned doesn't mean that the district does not have poor performing teachers sitting in classrooms," Deasy said in cross-examination.  "The second issue is that this is the snapshot of that day.  They had been in classrooms for years."

In his first day of testimony, Deasy estimated that the cost of firing a teacher was $350,000 on average and, in some cases involving egregious conduct, more.  "That's six teachers that could have been brought in to reduce class size (at the $350,000)," he said.

Deasy's conclusion:  "When you follow the law, an unfortunate byproduct of following the law is, in my opinion, the discrimination of youth having to be placed in front of an ineffective teacher."

In cross-examination, the defense referred often to Deasy's increased action to identify poor teachers and move them out of the school district as evidence that the system was working.

"Well-run districts," said the lawyer for the teachers' unions, "are able to fire ineffective teachers."

Quantifying the Differences between Effective and Ineffective Teachers

Raj Chetty, another plaintiffs' witness, set out to quantify the effects bad teachers have on students and their futures.  Chetty is a Harvard economist and affiliated with the National Bureau of Economic Research.

Chetty admitted to being a big-data guy who tried "to bring a scientific approach to public policy decision-making."

Chetty and others data-mined the effects of teachers on 2.5 million New York children using extraordinarily detailed school data from the New York district, 18 million test results and IRS filings regarding student college placements from 1989 to 2009.  After a long, long discussion of methods, Chetty said there were significant differences in learning (he called it value-added) for students in average teachers' classrooms and those of the bottom five percent of teachers. He said students learned significantly, quantifiably less when taught by poor teachers.

Chetty said that, had he been asked before his research project began in 2009, whether he could have identified the differences with such clarity, "I would be skeptical..... (I would have thought it) would be hard to isolate the effectiveness of teachers.  I was wrong."

Chetty is an egghead.  We may think he is stretching his analysis too far, but remember that he examined classroom-by-classroom results and the college placements of New York students over a 20-year period.

Here is his conclusion:

"If we replace an ineffective teacher with a teacher of average quality (in all cases), the impacts would be on the same order as ending the financial crisis again and again and again, year after year.  It would be a dramatic effect on the American economy in the long run."

Let's note also that Chetty's research has been taken up and mentioned by the federal Department of Education and the Obama administration on several occasions.



Discussion


Diane Ravitch, America's esteemed education historian, has not been called to testify in the Vergara case.  But she has been for decades a zealous proponent of educational improvement.  In 2000, she was involved in the formation of the National Council on Teacher Quality (NCTQ), a conservative group.  At that point, she said, teacher training institutions "were too touchy-feely, too concerned about self-esteem and social justice and not concerned enough with basic skills and academics."

Since that time, there has been a huge shift in schools toward what Ravitch calls "data-driven assessment."  In 2012, Ravitch broke with the NCTQ in a column in the Washington Post, saying this:

"Now, to be candid, I am fed up with our nation's obsession with data-driven instruction....I fear that they are pushing data-worship and data-mania of a sort that will cause teaching to the test, narrowing of the curriculum and other negative behaviors (like cheating).  I don't think any of this will lead to the improvement of education.  It will lead to higher test scores, but it will undermine genuine education ... a love of learning, a readiness to immerse oneself in study of a subject, an engagement with ideas, a willingness to ask questions and to take risks.  I don't know how to assess the qualities I respect, but I feel certain that there is no standardized, data-driven instruction that will produce what I respect."

The Question

And so here we are.  Students are being evaluated on test scores, and teachers who do not deliver regular improvements in test scores are being found faulting.  These teachers are challenging whether they should be held to account for student failures on quite basic measures.  Students from poor and minority backgrounds are claiming a right to achieve at least this much.

Meanwhile, educational experts like Ravitch are urging us to set the bar much higher.