roland fryer returns

Study: $75M teacher pay initiative did not improve achievement

New York City’s heralded $75 million experiment in teacher incentive pay — deemed “transcendent” when it was announced in 2007 — did not increase student achievement at all, a new study by the Harvard economist Roland Fryer concludes.

“If anything,” Fryer writes of schools that participated in the program, “student achievement declined.” Fryer and his team used state math and English test scores as the main indicator of academic achievement.

Schools could distribute the bonus money based on individual teachers' results, but most did not. Most teachers received the average bonus of $3,000.

The program, which was first funded by private foundations and then by taxpayer dollars, also had no impact on teacher behaviors that researchers measured. These included whether teachers stayed at their schools or in the city school district and how teachers described their job satisfaction and school quality in a survey.

The program had only a “negligible” effect on a list of other measures that includes student attendance, behavioral problems, Regents exam scores, and high school graduation rates, the study found.

The experiment targeted 200 high-need schools and 20,000 teachers between the 2007-2008 and 2009-2010 school years. The Bloomberg administration quietly discontinued it last year, turning back on the mayor’s early vow to expand the program quickly.

The program handed out bonuses based on the schools’ results on the city’s progress report cards. The report cards grade schools based primarily on how much progress they make in improving students’ state test scores. A so-called “compensation team” at each school decided how to distribute the money — a maximum of $3,000 per teachers union member, if the school completely met its target, and $1,500 per union member if the school improved its report card score by 75%.

The deal was seen as a landmark in 2007 when Mayor Bloomberg announced it with then-United Federation of Teachers President Randi Weingarten by his side. “I am a capitalist, and I am in favor of incentives for individual people,” Bloomberg said then, while Weingarten emphasized that schools could decide to distribute bonuses evenly among educators. She called the program “transcendent.”

In his study, published as a National Bureau of Economic Research working paper, Fryer writes that researchers were surprised to see that schools that won bonuses overwhelmingly decided to distribute the cash fairly evenly among teachers. More than 80 percent of schools that won bonuses gave the same dollar amount to almost all of the eligible educators.

Researchers were also surprised to find that middle school students actually seemed to be worse off. After three years attending schools involved in the project, middle school students’ math and English test scores declined by a statistically significant amount compared to students attending similar schools that were not part of the project.

The study adds to a research literature on teacher incentive pay that is decidedly more lukewarm than much of the popular conversation about teacher pay. Fryer, himself a strong early advocate of experimenting with financial incentives to improve student achievement, calls the literature “ambivalent.” While programs in developing countries such as India and Kenya have had positive effects, few teacher incentive pay efforts in the United States have been deemed effective.

Almost all schools gave nearly all of their teachers the same sized bonus.

Nevertheless, a person’s position on teacher merit pay has become a litmus test for her reform credentials in many education circles. During his campaign, President Obama used his support for merit pay — traditionally scorned by teachers unions — as evidence that he was willing to challenge traditional Democratic Party thinking. Now, the Obama administration has boosted support for the Teacher Incentive Fund, a program that funds local experiments in incentive pay.

What explains the discrepancy between programs in the U.S. and elsewhere? Fryer rejects several explanations. He argues that the $3,000 bonus (just 4 percent of the average annual teacher salary in the program) was not too small to make a difference, citing examples of effective programs in India and Kenya that gave out bonuses that were an even smaller proportion of teachers’ salaries. He also rejects the possibility that schools’ decisions to use group, rather than individual, incentives was the problem, citing a 2002 study of a program in Israel that used group incentives.

Instead, he says the challenge is that American plans aren’t clear about what teachers can do to receive the reward. In New York City, the bonuses didn’t come simply if students’ test scores rose; the test scores had to rise in comparison to a group of similar schools. So did other measures considered by the city report card, including the surveys that ask students, teachers, and parents for subjective opinions about schools.

Fryer argues that the complexity made it “difficult, if not impossible, for teachers to know how much effort they should exert or how that effort influences student achievement.”

money matters

Report: Trump education budget would create a Race to the Top for school choice

PHOTO: Official White House Photo by Shealah Craighead
President Donald Trump and U.S. Secretary of Education Betsy DeVos participate in a tour of Saint Andrews Catholic in Orlando, Florida.

The Trump administration appears to be going ahead with a $1 billion effort to push districts to allow school choice, according to a report in the Washington Post.

The newspaper obtained what appears to be an advance version of the administration’s education budget, set for release May 23. The budget documents reflect more than $10 billion in cuts, many of which were included in the budget proposal that came out in March, according to the Post’s report. They include cuts to after-school programs for poor students, teacher training, and more:

… a $15 million program that provides child care for low-income parents in college; a $27 million arts education program; two programs targeting Alaska Native and Native Hawaiian students, totaling $65 million; two international education and foreign language programs, $72 million; a $12 million program for gifted students; and $12 million for Special Olympics education programs.

Other programs would not be eliminated entirely, but would be cut significantly. Those include grants to states for career and technical education, which would lose $168 million, down 15 percent compared to current funding; adult basic literacy instruction, which would lose $96 million (down 16 percent); and Promise Neighborhoods, an Obama-era initiative meant to build networks of support for children in needy communities, which would lose $13 million (down 18 percent).

The documents also shed some light on how the administration plans to encourage school choice. The March proposal said the administration would spend $1 billion to encourage districts to switch to “student-based budgeting,” or letting funds flow to students rather than schools.

The approach is considered essential for school choice to thrive. Yet the mechanics of the Trump administration making it happen are far from obvious, as we reported in March:

There’s a hitch in the budget proposal: Federal law spells out exactly how Title I funds must be distributed, through funding formulas that sends money to schools with many poor students.

“I do not see a legal way to spend a billion dollars on an incentive for weighted student funding through Title I,” said Nora Gordon, an associate professor of public policy at Georgetown University. “I think that would have to be a new competitive program.”

There are good reasons for the Trump administration not to rush into creating a program in which states compete for new federal funds, though. … Creating a new program would open the administration to criticism of overreach — which the Obama administration faced when it used the Race to the Top competition to get states to adopt its priorities.

It’s unclear from the Post’s report how the Trump administration is handling Gordon’s concerns. But the Post reports that the administration wants to use a competitive grant program — which it’s calling Furthering Options for Children to Unlock Success, or FOCUS — to redistribute $1 billion in Title I funds for poor students. That means the administration decided that an Obama-style incentive program is worth the potential risks.

The administration’s budget request would have to be fulfilled by Congress, so whether any of the cuts or new programs come to pass is anyone’s guess. Things are not proceeding normally in Washington, D.C., right now.

By the numbers

After reshaping itself to combat declining interest, Teach For America reports a rise in applications

PHOTO: Kayleigh Skinner
Memphis corps members of Teach For America participate in a leadership summit in last August.

Teach for America says its application numbers jumped by a significant number this year, reversing a three-year trend of declining interest in the program.

The organization’s CEO said in a blog post this week that nearly 49,000 people applied for the 2017 program, which places college graduates in low-income schools across the country after summer training — up from just 37,000 applicants last year.

“After three years of declining recruitment, our application numbers spiked this year, and we’re in a good position to meet our goals for corps size, maintaining the same high bar for admission that we always have,” Elisa Villanueva Beard wrote. The post was reported by Politico on Wednesday.

The news comes after significant shake-ups at the organization. One of TFA’s leaders left in late 2015, and the organization slashed its national staff by 15 percent last year. As applications fell over the last several years, it downsized in places like New York City and Memphis, decentralized its operations, and shifted its focus to attracting a more diverse corps with deeper ties to the locations where the program places new teachers. 

This year’s application numbers are still down from 2013, when 57,000 people applied for a position. But Villanueva Beard said the changes were working, and that “slightly more than half of 2017 applicants identify as a person of color.”