Catalyst Caucus

It is quite apparent that nationally there is a growing dissatisfaction with how states and school districts evaluate their teachers. Most district evaluation systems fail to identify weaker teachers and teaching and fail to provide useful information to teachers on how to improve their classroom practice. In a number of districts, teacher evaluation is still conducted by principals with little or no formal training using observational tools or “checklists” that are outdated and do not effectively assess classroom instruction.

These weaknesses are one reason the federal Race to the Top competition is encouraging states like Illinois to improve their teacher evaluation systems by, among other things, incorporating data on student performance into the assessments of individual teachers. New legislation in Illinois—the Performance Evaluation Reform Act of 2010—requires that standards-based teacher evaluation systems with a student achievement indicator be implemented throughout the state. Other states like Florida are moving even more aggressively to tie teacher tenure and compensation to performance-based assessments of teaching.

Illinois and other states are to be commended for taking on their broken teacher evaluation systems, yet the path forward on how to fix these systems is less than clear. Measuring teacher “value added” is more difficult than many advocates suggest and many states—including Illinois—lack the data systems that can connect individual student achievement gains to specific teachers and classrooms. Yet while many states are moving to link teacher evaluation to student performance, the majority will still continue to base half or more of teacher evaluations on other measures of effectiveness besides student test scores.

What might these other measures look like and how useful are they? In Chicago we are beginning to learn just that. The Excellence in Teaching Pilot, introduced in Chicago Public Schools (CPS) in 2008, represents a particularly ambitious district-led effort to improve teacher evaluation. The pilot has adapted the Charlotte Danielson Framework for Teaching to create a new evaluation tool that principals can use to assess teaching practice. The pilot evaluation tool is the proposed replacement to a simple checklist that has been used in CPS for the last 30 years. As part of the pilot, extensive training and supports has been provided for participating principals in using the new evaluation tool.

Over that last 18 months, my colleagues at the Consortium on Chicago School Research (CCSR) have been studying the Excellence in Teaching pilot, which is currently being implemented in 44 Chicago elementary schools. In the first phase of the evaluation, CCSR has focused much of its attention on whether the new evaluation tool is a reliable and valid instrument for assessing teaching. We were able to get at this by employing a study design that uses “matched” observations of where external observers and school administrators conduct classroom observations at the same time and then rate the teachers independently.

Among the study’s initial findings are that the new evaluation tool can reliably identify low-quality teaching—with principals and observers generally rating teachers in similar ways. However, to fully realize the potential of this new tool, ongoing training and support for principals on the use of tool is crucial. CPS provided ongoing professional development and support for principals in the first year of the pilot; yet a number of principals still struggled to rate some areas of instruction (e.g. student engagement in learning) consistently. Principal turnover and the need for extensive training as the initiative expands to all schools pose particular challenges going forward for CPS.

Lastly, there are significant future challenges in using the new evaluation tool for high-stakes decisions. Inconsistencies in the way that principals rate some areas of instruction and differences in severity across principals pose major challenges for using the tool to make consequential decision about teachers. For example, having a principal who is a severe rater might have detrimental effects on the careers of borderline teachers in that school. On the other hand, lenient principals might keep teachers who should otherwise be removed due to low performance. These ongoing difficulties suggests that policy makers in Chicago and elsewhere should move cautiously as they implement new evaluation systems and ensure the availability of proper support for teachers and principals in implementing the Danielson-based tool and other new measures of teacher effectiveness.

As a student at Roberto Clemente, I can say that teachers should not be evaluated on how well their students do on a test or on their grades. Not every student learns the same way and has the self motivation to do good in school like other students. Personally, I feel that some of my teachers are great but my grade and test scores does not reflect them as a teacher. A good teacher is one that offers up their extra time to help you and is one that encourages and gives you advice whenever they notice there is something wrong. Part of what teachers should be evaluated on is how much their students trust them because building trust with a student is what will help them stay in school and more committed to class.

Diana Carreno, 12th grade
Roberte Clemente High School

Just as teaching itself is both art and science, so too is the process of evaluating teachers. While there are certainly exceptions, the system has not worked effectively either to give teachers good feedback on how they can improve classroom practice, to identify critical professional development needs, to celebrate exceptional teachers, or to identify weak teachers. The notion that a teacher's evaluation (or that of a principal) should depend in some way on his or her ability to impact student learning makes intuitive sense. At the same time, and as Jackson Potter, Chris and several of the student participants point out, understanding student growth and learning is complicated and doesn't necessarily take into account all of the things we want and expect teachers to accomplish in the classroom. This reality prompts a number of observations.
First, while most agree there are no easy answers for how to measure student growth (current standardized tests were not meant for this purpose, and plenty of courses and grades do not even take such tests), it is healthy for teachers and administrators to be talking about what they expect of their students and classrooms, and how they will know if they've made progress against those goals. The changes underway are prompting these conversations, and that is a positive shift in thinking and discussion.
Second, no evaluation system should depend entirely on measures of student academic growth. As Jackson and several students point out, outstanding teachers are willing to take extra time with students, to motivate them beyond their studies, and to help shape their character. Thoughtful means to evaluate classroom practice (such as the Danielson Framework mentioned above) can and should be used to understand teacher performance and to help craft useful feedback.
Finally, while I feel differently about the value and use of competitive grants (and would point out that competitive Race to the Top dollars come only on top of signficant stimulus dollars for all states and schools), I share Jackson's view that we are overdue for a more sophisticated way of evaluating school success. We should look at a broad array of information to understand what is working and what is not at a school, and to ensure that schools do not narrowly focus on certain test results to the exclusion of building a climate and culture that stimulates and nurtures students and professionals alike, or at the cost of ensuring children have access to a rich array of courses and experiences.