Pay-4-Performance: Individual vs. Group Incentives

Schools are identified as failures under No Child Left Behind. Turning around a school may include firing the staff or even closing the school. In short, the school is the unit for punishing failure.

And that is smart because when it comes to determining failure policymakers understand that schools are collectives of adults and children where structures and group cultures have an enormous influence on what individual teachers and students do. They also recognize that schools are interdependent units of teachers–the 4th grade teacher depends upon the kindergarten through third grade teachers to prepare her students for the work that her 10 year-olds will be doing. Similarly, high school physics teachers depend upon 9th grade math and science teachers to have thoroughly prepared their students for their courses.

If it is policy smart to see schools as units when failure and turnarounds are concerned, then why the passion and hullabaloo for paying individual teachers based on test scores rather than rewarding the entire school?

Much of the answer, of course, goes to the love affair that efficiency-driven policymakers have had with the success of business approaches since the late-1970s. Then, Japanese business approaches incorporating quality control, restructuring, just-in-time inventories, and other efficiencies were imported into U.S. industries. The spectacular business boom between the late 1980s- through the 1990s (except for the brief recession that threw President H.W. Bush out of office and elevated Arkansas governor Bill Clinton to the presidency) convinced many policymakers and political leaders that successful business approaches could (and should) be easily imported into school improvement. If schools had to compete like businesses do, innovation and efficiencies would surge and improvement would spread. Enter expanded parental choice and charters. If districts would move away from single salary schedules and embrace rewarding successful teachers and dumping unsuccessful ones–as businesses have to do every day with their employees–then students would profit from the best teachers and avoid the worst. Enter pay-4-performance plans. No Child Left Behind becomes law in 2002 providing test scores as the single most important basis for identifying failing schools. The technology for identifying those teachers within each school whose students did well and those who did poorly was already available.

The idea that individual teachers could be determined “effective” based on their students test scores a la the Tennessee value-added assessment studies, has produced a cascade of plans (also see Podgursky and Springer-TeacherPerformancePay) to evaluate teachers, identify “effective” ones, and then pay those a premium based on their performance. These schemes have gained a great deal of traction since the recession of 2008, the joint Bush/Obama education agendas, and recent efforts to cut education spending.

In education, however, political and policy consensus often ignores research findings that challenge that consensus. There has been a growing consistency of findings, for example, on the failures of teacher merit pay since the 1920s or offering individual incentives. See here (fulltext-8) and here. There is also an emerging literature that piece rate payments (individual incentives) in manufacturing is less effective than group incentives (see PDF–w16540). Even more striking is the evidence from developing countries like Kenya (w9671) and India (w15323) where the kinds of research–experimental/ control studies with rigorous randomized assignment of schools and teachers that policymakers advertise as the gold standard and called for in NCLB–found that group incentives at the school paid off far more than individual ones given to teachers.

Of course, such evidence drawn from from historical studies of merit pay and research done in other countries–even when carried out by economists who policymakers prize–have little influence on decision-makers and their allied donors precisely because they cut holes in the current consensus of paying individual “effective” teachers more than those identified as less “effective.”

Research studies past and present, in education and in business, challenge this group-think among corporate leaders, the past three U.S. Presidents, and state governors. Seldom, however, do these studies ever show up in public debates. Their absence from policy talk and action testifies to the weakness of research findings–even when contributed by economists–on educational decisions except on those occasions when the decision already has been made and then studies are trotted out to justify the decision.

If research findings that undermine the current “wisdom” are shrugged off, what about policy consistency? Today, those schools identified as failing are punished by firing staff, being restructured, converted into charter schools or shut down. There is a practical wisdom to recognizing that teachers and student work interdependently within school structures and cultures to produce failure. Yet if the unit is the school for determining failure and then punishment, it seems minimally consistent that the school, not individual teachers or principals, should also be the unit for rewards–beyond a plaque or flag.



Filed under school reform policies

6 responses to “Pay-4-Performance: Individual vs. Group Incentives

  1. John Randal Guptill

    We had low performing kids coming in due to re-districting that threatened to put our school in “low performing” status. We identified less than 20 kids who were in trouble with the testing program. Teachers volunteered half their planning time to tutor these kids. Long story short, these kids did the best they ever had on the state mandated tests, we beat the target score the state assigned us and the faculty got a bonus. Bad news was the system was fixed. If you met your goals, the state would bump them higher and higher every year until they couldn’t be met, then your school was in trouble again.

  2. Clearly true that teacher effectiveness (no quotes required) can be measured by value-added analysis. And clearly true also that your critique of the policy response of individual pay-for-performance incentives is true also. This response assumes that most teachers are just not trying and could improve significantly with just a little more effort, an belief for which there little, if any, data. The more accurate belief would appear to be that most teachers are doing about the best they know how under all the circumstances. So, while there is little evidence that schools impact teacher performance (generally, teachers do not become significantly more or less effective with students in a particular range of performance just through changing schools), the school is still the best unit in which to focus on individual change.

    Change is hard. Can be exciting, but definitely requires work. Most of us undertake and sustain the effort necessary to change in ways that move us toward our goals in life better when we work with others. As Chris Peterson says, “Other people matter.”

    However, policy leaders aren’t the only ones who lack trust in teachers. School and district leadership also don’t trust teachers. That’s why teacher-led-instructional improvement is so hard to implement properly as a professional development approach. Central office administrators and too many principals immediately start to change the key element – “teacher-led” to “keep teachers from screwing it up.”

    “Other people matter.”

  3. Larry,
    Two things. The most successful project I ever worked on which involved working directly with groups of full time teachers, (and there have been many!) involved my being in a position to pay them cash for the additional work they did. Second, I think your post raises a really interesting challenge to the policy makers which is how do you design whole school incentives that genuinely incentivise the whole school community.

    And for when you have a spare moment, a weighty tome of World Bank research that deals very specifically with the issue of teacher incentives which I just happened to be reading today!

  4. Robert Tellander

    Socially, a school is a plural institution that tries to educate individuals.

    More properly perceived, we do not educate one at a time, but the students themselves create a learning context. Teachers orchestrate learning and that does not succeed as rote or in unison. Consequently, your economic argument applies, but it is better understood as a social reality not a better economic option.

  5. Pingback: links for 2011-04-17 : DrAlb

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s