Wednesday, October 23, 2013

DC Teacher Incentives

Michelle Rhee's tenure as Washington DC's Chancellor of Public Schools was controversial mostly because she instituted reforms designed to hold teachers accountable for classroom performance. This episode provides the backdrop for studying the role of high-powered incentives linked to multiple measures of teacher performance. The effectiveness of one of these reforms have recently been analyzed by Thomas Dee and James Wyckoff in their paper "Incentives, Selection, and Teacher Performance: Evidence from IMPACT." So how did it do? From the abstract:
Our RD [Regression Discontinuity] results indicate that dismissal threats increased the voluntary attrition of low-performing teachers by 11 percentage points (i.e., more than 50 percent) and improved the performance of teachers who remained by 0.27 of a teacher-level standard deviation. We also find evidence that financial incentives further improved the performance of high-performing teachers (effect size = 0.24).

So screening mitigated both the adverse selection and the moral hazard problems and not by small amounts.


  1. Devil's advocate: The tough part with the apparent improvement in this scenario is incentive alignment. No measurement tool is perfect, resulting in rational-actor teachers manipulating the measurement in order to secure the financial incentives for themselves.

  2. I'm interested to know what constitutes "low performing" in this case. Usually performance is unfortunately based on poorly designed, state mandated tests which do not accurately capture a student's progress or success in the context in which they are in. For example, a freshman may fail the test but has really made great strides by improving 2 reading levels, though he needs more time to be at grade level. His teacher may have done a fabulous job of inspiring him to continue learning even after he leaves her classroom but the state is saying her scores are low so let us threaten her for dismissal. She leaves, as many teachers are nowadays because they are fed up with the ridiculous policies and bureaucracy which are actually hindering student learning and creativity, and the school ends up losing a great teacher to private or charter schools. Check out this New York independent school's approach to education

    1. It would be great to look at student performance on a more individual basis. I agree that individuals learn differently and at his/her own pace. However, how would we rate something that can be seen as so subjective?

      Went to the blueschool website, and found this on there: "Teachers ask that they step out of their comfort zones to experience another view." In my opinion, this may be one of the most effective learning tools. Especially at a young age, if children can learn to understand both sides of a certain view or position, then they become more well-rounded and smarter individuals.

  3. I agree with Leiya's assessment of this post. Although the performance of a student may be an indicator of the quality of education, it is important to realize that there are several social factors that can also have an impact on the performance of a student. School districts across the country should incorporate more than one evaluation method to properly assess teachers. For example, the state of Tennessee looks at a variety of factors when determining the effectiveness of a teacher. Over 50% of the performance evaluation is centered around observations by administration and by district peers. It is important to also note that performance of students can vary by region, school district, and even area.

  4. Another thing these standards and practices do is increase the incentive for teachers to cheat, which has been a growing problem in DC and around the country:

    The practice we're trying to incentivize here is student learning and instead we've incentivized cheating. The portion of teacher assessment dependent upon objective classroom observation should increase.