As a master teacher at the San Antonio Independent School District, Michelle Olivarri gets a $15,000 stipend to teach at a school with a history of low student outcomes.
During a reading lesson in her third grade classroom earlier this semester, sounds of excitement blended with energetic music from a computer game on the parts of a story.
“I’m in second,” shouted one student after filling in his answer on a tablet.
“I’m in fourth,” said another. “Ms. Olivarri, I was on fire!”
Many of Olivarri’s students began the school year significantly below grade level. Because the district looks at standardized test scores when it hires master teachers, she’s not sure the district would have chosen her for the performance-pay program if she had applied using this year’s test scores.
San Antonio ISD is one of a handful of Texas school districts that pays some teachers more based on their performance — a program that could get state funding if the school finance overhaul bill HB 3 becomes law.
San Antonio’s version is intended to help recruit and retain high-quality teachers in the schools that need them most, but the criteria the district uses to decide which teachers qualify for the higher pay could favor teachers that come from more affluent schools.
One way the district picks master teachers is by looking at how their students performed on the state’s standardized test.
“The year before was higher than the last year, and then this year seems like it's, you know, there's, there's more of a struggle with this group. And, you know, I’m doing what I know I need to do,” Olivarri said.
“The teachers that have the low group like I'm having this year now is it fair to them? Probably not.”
District leaders said the goal of the Master Teacher Program is to help students catch up academically. Applicants can qualify for the stipend is if their students show measurable improvement on their test scores, but when that data isn’t available, teachers can only qualify based on their students’ performance that year alone. That metric strongly favors teachers whose students come into the classroom already on grade level.
It’s one reason why many teacher unions — and some parent groups — oppose performance-based pay.
Louis Malfaro, the president of the Texas chapter of the American Federation of Teachers, said he’s concerned it could discourage teachers from working in struggling schools.
“That in fact what's going to happen is they're going to be more perverse disincentives for teachers not to go to work in those schools,” Malfaro said.
Research shows that, on average, students from low-income families score lower on standardized tests than students from wealthier families. A Texas Public Radio analysis of state accountability data found that same correlation.
However, San Antonio ISD Deputy Superintendent Matthew Weber still thinks it’s important to use both metrics: student improvement and performance. He said he doesn’t think the district’s criteria runs the risk of favoring teachers who currently work at more affluent schools.
Weber said test results aren’t SAISD’s only criteria. It also looks at teacher evaluations and requires two years of experience. He oversaw the launch of the Master Teacher Program two years ago, and he said the district is open to changing its selection qualifications if the evidence shows it’s needed.
Still, he acknowledges that teaching in a high-poverty school can be challenging — and that teachers from more affluent schools have had a hard time adjusting when they come to SAISD.
“Even if they have super evaluations and great scores, when they come into a high-poverty school they are not being as successful as the ones that we may have internally or the ones that we’re getting from surrounding districts,” Weber said.
||More to the Story: A Rubric of Student Test Results SAISD Prioritizes When Hiring Master Teachers ||
Researchers who study performance-based pay agree that having multiple ways of measuring teacher quality can help make them fairer and more effective.
“It's always really important that we use a multiple measure outcome,” said Matthew Springer, a professor of education evaluation and public policy at the University of North Carolina at Chapel Hill. “And by that I mean that we're taking into account student test scores; that we're taking into account classroom observations; that we're taking into account these other more holistic assessments that we have of teacher performance.”
Springer is a co-author on a recent analysis of a decade’s worth of studies on teacher performance pay. It found that U.S. programs added about three weeks of learning.
But as long as they’re tied to test scores, the potential to favor teachers with students from middle and upper class families remains.
“It does have that potential,” said Dan Goldhaber, director of the Center for Education Data and Research at the University of Washington. “But … think about not letting the perfect be the enemy of the good.”
Goldhaber said even an imperfect merit pay system may be better than the usual way districts pay teachers — small raises based entirely on years of experience.
“As a member of society and a taxpayer I care about the fairness to public employees, but I think I care more about whether the system that is adopted is beneficial to students,” Goldhaber said. “Certainly as a parent I do.”
If the House version of merit pay becomes state law, it would have several ways of measuring teacher performance, like Springer and Goldhaber recommend. But as it stands now, one of those measures would most likely be a standardized test.