Value Added and the Teacher Incentive Fund

The Teacher Incentive Fund, created by Congress in 2006, provides grants to districts and schools to create performance-based compensation systems for teachers and administrators in high-needs schools. Under the program’s guidelines, grantees “must consider gains in student academic achievement” as part of the proposed compensation systems.

States, school districts or charter schools, often in partnership with nonprofit organizations, receive grants of from $100,000 to $10 million a year for up to five years. Congress in its fiscal 2010 budget increased funding for the program from $97 million to $400 million, so many more awards are expected.

In an analysis of the 34 grants awarded during the Teacher Incentive Fund’s two initial rounds, Robert H. Meyer and Michael S. Christian found that half proposed the use of value-added models for measuring teacher effectiveness. Most of these were medium to large school districts, and all but one (Dallas) relied on an outside organization to conduct the value-added analyses.

The rest of the grantees proposed simpler measures they could calculate in house. These included measuring teacher perform-ance through the use of gains in average achievement; the number of students who reached the proficient level of achievement; the movement from one level of proficiency to another; or some hybrid combining these measures. One grantee proposed using gains based on individualized achievement plans.

Meyer and Christian, who did the study for Vanderbilt University’s National Center for Performance Incentives, concluded that the value-added measures, despite their statistical complexity, were superior to the others as measures of school and teacher performance. The value-added scores were more accurate than average achievement gains, they found, and the others posed problems that could result in erroneous measures of school and teacher performance.

“The value-added model is highly attractive from a technical perspective,” they conclude. “A value-added model, particularly one that is enhanced and customized to local conditions, is capable of handling numerous real-world problems that might otherwise threaten the validity of a simpler model or indicator.”

— Robert Rothman