Note: Paul Manna, a professor at William & Mary, is guest-posting this week while Rick Hess is on a consulting project in the Republic of Georgia.

Let me wrap up this week by looking closely at one element of Race to the Top (RTT) that has prompted much discussion: teacher evaluations. Specifically, I’m thinking about part D(2) of the RTT criteria, which focuses on “Improving teacher and principal effectiveness based on performance.” (See p. 59821 in this document or p. 19504 in this one.) The RTT vision is for teachers (let’s leave aside principals for now) to face annual evaluations of their performance with student achievement gains carrying major weight in those judgments.

The majority of the debating points I’ve read on this issue have tended to focus on whether, in principle, teacher evaluation should or should not include a major component focusing on student achievement as measured by state tests. What’s been missing or understated in the discussion, I think, is exactly how either side envisions its preferred evaluation method happening in practice. Again, it’s the theme I’ve been pushing this week: Implementation, implementation, implementation. Specifically, let’s consider a key variable--time--and why it requires serious discussion in the teacher evaluation debate.

While the use of student performance data to evaluate teachers has an intrinsically appealing quality, it is important to remember that in the NCLB era states and their testing contractors have been notoriously bad at producing results in a timely manner. The vast majority of states, for example, have been unable to issue schools’ final adequate yearly progress (AYP) judgments until well into the summer, if not into the following school year. In the future, the ability to turn around test results in a timely manner will be especially crucial if annual employment decisions or teacher salaries and bonuses will be tied to these test results. Having the final numbers from spring testing ready in late August or September, even, simply won’t cut it. Compounding the challenge is that the reporting tasks become still more complicated when teacher evaluations, not just school evaluations, are at stake. Do the math for a minute. It was hard enough for states to compute performance results (i.e., AYP) for the nearly 100,000 schools in the country each year. Add to that the future need to generate results for some relatively large fraction of the nation’s three million teachers and tens of thousands of principals and the data management task becomes truly daunting given the current (and likely near future) state of the art.

The other time element meriting more attention in the debate is simply the amount of time needed to do rigorous evaluations. Presently, teacher evaluation is a pretty anemic process in most districts. Formal evaluations, if occurring at all, typically are cursory exercises that involve principals and teachers having a meeting, jointly completing evaluation forms about the teacher’s performance, and both signing off on the document. Serious classroom observations of teachers or teacher peer review tend to occur about as often as French goals in this year’s World Cup. This is not because teachers and principals are necessarily uninterested in improving their practice. Rather, the structure of their jobs tends to crowd out space for such work. As political scientist James Q. Wilson has explained, immediate circumstances (e.g., principals focusing on school discipline and teachers prepping their lessons and grading papers) tend to take priority over tasks requiring longer-term effort and planning. Like all serious work, meaningful evaluation requires what Max Weber once called the “slow drilling of hard boards.” Cutting corners simply won’t get the job done. (Hat tip to my colleague Simon Stow for reminding me of Weber’s wisdom.)

So in the end here I am left with lingering questions for both sides of the RTT teacher evaluation debate. To those who favor incorporating measures of student achievement gains: Why are you confident, given state track records, that we are anywhere near possessing the data systems required for timely (and obviously accurate) reporting of test results needed for annual teacher evaluations? To those who wish to see a more rigorous and multi-dimensional approach to evaluation that incorporates factors beyond test scores, I wonder: Where do you envision teachers and principals finding the time to perform the tasks that such robust evaluations would require? What current activities do you envision them shedding in order to create space to make the evaluation process a serious part of their work?

In closing, let me thank Rick again for allowing me to share some thoughts with you this week. If you’re interested, please feel free to be in touch (pmanna@wm.edu) or surf to my home page to check out my other work, including an excerpt from my forthcoming book on NCLB’s implementation called Collision Course: Federal Education Policy Meets State and Local Realities. Take care.

--Paul Manna