School of Ed profs brief Capitol Hill on education reform

At a Sept. 14 Capitol Hill briefing, School of Education professors Linda Darling-Hammond and Edward H. Haertel, among other leading education researchers, presented their findings on a central concern in federal and state policy  –  how teacher performance should be evaluated.

School of education professor Linda Darling-Hammond delivers the keynote address at an education forum in 2008. (Stanford Daily File Photo)

One popular type of teacher evaluation is the value-added method (VAM) system, which uses statistical methods to measure gains in student achievement. The VAM system measures changes in student test scores over time, while taking into account other factors that are found to influence achievement.

According to Haertel, who also serves as chair of the National Research Council Board on Testing and Assessment, one major flaw of the VAM system is that it gives policymakers the illusion of a “quick fix” for improving student performance by identifying teachers’ level of competency through their VAM scores.

“There’s a hope that [problems will be solved] by funding poor teachers that need to either be given some remedial assistance or discouraged from continuing to teach,” Haertel said. “I’ve not been surprised that these models have been caught on and have been so popular among policymakers.”

Surprisingly, the problem with the VAM system is not with its reliance of standardized testing scores, he continued.

“It’s not the tests, per se,” Haertel said. “It’s easy to say we need better tests, but that has been tried again and again. The tests we have are pretty good at what they do.”

Because the VAM system does not account for factors such as the amount of teacher infrastructural support, class sizes and individual student learning needs, an important side-effect of the system is an increase in competition between teachers to perform well. This side effect may result in teacher neglect of the population of students who may need the most help, such as low-income students from non-English-speaking backgrounds.

“We create incentives with these value-added systems with teachers who try to avoid students who might do poorly on the tests or students who might not make rapid gains in the test scores,” Haertel said.

The VAM system’s measurement of teacher effectiveness has proven unstable and unreliable. One study showed that out of teachers who scored in the bottom 20 percent of rankings in one year, only 20 to 30 percent had similar ratings the next year. These results suggest that teacher evaluations should focus more on measuring and expanding professional development.

“We need both professional development and accountability, and right now, the equation is out of balance,” Haertel said. “We have much more focus on accountability and much less on giving teachers the resources they need to develop skills and do a better job of accomplishing their work.”

Darling-Hammond proposed alternatives to the VAM system, calling for performance-based assessments and standards-based evaluations. This alternative system will combine evaluations on different fronts in different mediums into a portfolio that reliably tracks the teacher’s growth in his or her teaching craft.

“It’s a pretty big job to assemble that evidence, which has to be scored by trained evaluators to produce a reliable score,” Darling-Hammond said. “That kind of evaluation can be done at certain key junctures, such as when you first enter the teaching profession, during tenure-based decisions, at evaluations where teachers can become mentor teachers  .  .  .  it is a very time-consuming and rigorous process.”

Despite the cost of implementing a more stringent teacher evaluation system, Darling-Hammond is optimistic that policymakers will be more enthusiastic about seeking teacher evaluation assessments that are more reflective of teacher effectiveness.

“There’s an appetite in investing [in] teacher evaluation,” Darling-Hammond said. “These value-added models are themselves quite costly, though they don’t actually end up being very reliable, stable or valid.”

“If you want to improve teaching, making investments in these programs that provide solid evaluation and more teacher assistance, it will probably prove to be less expensive and more productive,” she added.