Merit-pay study showed no gains

Performance pay for teachers didn’t boost student achievement, according to a Mathematica study of the Chicago Teacher Advancement Program. The first two years of the pilot also showed no improvement in teacher retention at participating schools.

Education Secretary Arne Duncan was running Chicago schools when TAP was started, with support from the teachers’ union, and his Race to the Top is pushing states to adopt performance-based pay schemes.

Nobody knows why TAP had no effect, reports Education Week. However, Chicago changed the TAP model, spreading bonus money among teachers, principal and staff instead of just teachers.

Because of problems with obtaining student-growth data linked to individual teachers, Chicago also paid bonuses based on schoolwide, rather than classroom achievement growth. The National Institute for Excellence recommends that at least 30 percent of bonus pay be based on the results of classroom measures of student growth.

Chicago also paid smaller bonuses than recommended — an average of $1,100 in the first year and $2,600 for  teachers in schools in their second year of TAP.

The federal Teacher Incentive Fund, which supports performance pay pilots, tells applicants “that average bonus payouts for educators should be ‘substantial,’ perhaps 5 percent of the average teacher salary, and that top-performing educators should earn far beyond that amount, perhaps three times as much,” reports Ed Week. Chicago’s plan may have been too diluted to make a difference.

About Joanne


  1. Bad Science says:

    The inertia that has developed over time that allows around 50% of students to fail to graduate in Chicago is the reason this didn’t appear to work. It will take longer than two years for merit pay to show positive results, and the only current alternative is ever increasing failure rates.

  2. Merit pay will have little effect on overall system performance (as measured by standardized tests, graduation rates, or college acceptance rate) so long as insiders get to define “merit”. See Axelrod, The Evolution of Cooperation. or an economics dictionary on “regulatory capture”.

    As ever, “What works?” is an empirical question which only an experiment can answer. In the education industry, this means a competitive market in education services.

  3. Michael E. Lopez says:

    The idea of “merit pay” being something like 4-5% of salary is laughable.

    The best (or at least best-paid) attorneys make FIVE TIMES what the average attorney makes.

    The best doctors make FIVE TIMES what the average doctor makes.

    Etc. Etc. Etc.

    Don’t expect any gains from being willing to throw around chump change.

    Push differential salaries up into the low $170,000’s and we’ll start, over time, to see some changes.

  4. Whyever would someone think that 2 years is sufficient time for better instruction to show up in test scores? I would think that it would take 4-5 years AT A MINIMUM. And I’ll betcha that the researchers only looked at average improvements, too. You really need to go one by one, i.e., look at each student, and see if there is any improvement in that one student. You get a + for every improved score, an = for every score that stays the same, and a – for every score that goes down. That way you can look at aggregate data in a number of different ways.

    So even if the average test scores stayed the same, you could tell if more students than not were improving.

  5. Eric Kendall says:

    “Nobody knows why TAP had no effect, reports Education Week.” Well, here’s one reason why it didn’t work: “Chicago changed the TAP model, spreading bonus money among teachers, principal and staff instead of just teachers.”

    I don’t expect teachers are going to strive very hard for “merit pay” if they have to share the financial rewards they’ve earned with administration and support staff. Clearly, Duncan and the teacher’s union either simply didn’t understand the whole point about establishing individual incentives, or they don’t want to stray from an essentially collectivist approach to running the educational establishment.


  1. […] This post was mentioned on Twitter by Miguel Guhlin, kriley19. kriley19 said: Joanne Jacobs: Merit-pay study showed no gains Full […]