In 2001, when Patricia Hopkins became superintendent of the Five Town CSD and Maine School Administrative District #28 in Camden and Rockport, one of her first tasks was to review summative evaluations of all the teachers in the two districts. What she discovered troubled but did not surprise her. As she read through the evaluations, she found that many were full of "valentines"—her word for vague, meaningless praise—and largely devoid of constructive criticism or concrete feedback. Hopkins believed that teacher evaluation held great potential to improve instruction, so she set out to "eliminate the valentines" by strengthening the culture and structures supporting teacher evaluation in district schools.
In recent years, the spotlight on teacher evaluation has intensified. Given teachers' effect on student learning and achievement, practitioners, policymakers, and researchers have all called for boosting the rigor and quality of teacher evaluation. The track record for evaluation, however, flies in the face of the increasing consensus that teacher evaluation could play an important part in improving teaching and learning. My 2009 review of research on teacher evaluation showed that, on the whole, teacher evaluation has not substantially improved instruction.
During the last wave of efforts to strengthen teacher evaluation, in the 1980s, most initiatives died on the vine. This time around, however, there may be cause for more optimism. Key changes in the education world may make it easier for broad-scale improvement efforts to take hold. Moreover, some schools and districts, like those Hopkins leads, have already taken steps to more tightly link teacher evaluation with instructional improvement and increased student learning and to implement real consequences for those who perform superbly—and for other teachers who perform poorly.
All Above Average
Time and again, analyses of summative evaluation ratings of teachers show that the vast majority of teachers in any school, district, or state are rated above—sometimes well above—average (for a summary of the research, see Donaldson, 2009). Although it is possible that all teachers are above average in some schools, there is generally more variation in teacher effectiveness within schools than between them. Thus, any school—low performing or high performing, wealthy suburban or underresourced urban—is likely to employ more underperforming teachers than its evaluation ratings suggest (see, for example, Hanushek, Kain, O'Brien, & Rivkin, 2005). In fact, both principals and teachers believe that teachers are less effective than ratings indicate.
Inflated ratings of teachers reflect the following problems that seriously limit the extent to which evaluation could improve instruction and achievement:
- Poor evaluation instruments. Systems have tended to emphasize what can be measured, not necessarily what matters. Thus, evaluation instruments have traditionally required evaluators to look for things that they can easily check off (such as the neatness of bulletin boards) but that may not indicate high-quality instruction.
- Limited district guidance. Districts typically give little direction regarding what evaluators should look for. Instead of providing guidelines and rubrics about the substance of evaluations, districts are more likely to set out time lines and explain processes (Koppich & Showalter, 2008).
- Lack of evaluator time. Evaluators, usually school administrators, report having insufficient time to conduct thorough and accurate evaluations. As the reporting requirements for schools have increased, evaluators' time has become even scarcer.
- Lack of evaluator skill. Evaluators often lack specific knowledge about the content areas in which they evaluate teachers, especially at the secondary level. Moreover, professional development for evaluators is not frequent or comprehensive.
- Lack of evaluator will. Principals are not always held accountable for conducting rigorous evaluations. A "culture of nice" pervades schools, suppressing critical feedback and encouraging principals to rate all teachers above average.
- Absence of high-quality feedback for teachers. Even though teachers express a strong desire for more concrete, detailed feedback, evaluators generally do not provide it after their observations (New Teacher Project, 2009).
- Few consequences attached to evaluation. Because there is little variation in teachers' summative evaluation ratings, teachers who teach exceptionally well cannot be identified or rewarded. At the same time, it's difficult to identify and remediate or, if needed, dismiss those who struggle (New Teacher Project, 2009).
Overcoming the Challenges
Despite the deep, longstanding roots of these problems, the challenges might be easier to surmount than they appear. Currently, we know more about the links between teaching and learning than at any time in the past (Donovan & Pellegrino, 2003). We know, for example, that explicit instruction in the alphabetic principle is a key component of effective reading instruction (National Reading Panel, 2000; Snow, Burns, & Griffin, 1998). This knowledge enables evaluators to determine whether such instruction is occurring.
In addition, the teacher workforce is undergoing a massive transition as baby boomers retire and individuals in their 20s and 30s enter teaching. There is some indication that new teachers today differ from the retiring generation (Johnson & Project on the Next Generation of Teachers, 2004). Surveys suggest that they are more open to differential recognition and rewards than are their retiring counterparts. Finally, teachers unions, long perceived as a major barrier to the improvement of teacher evaluation, have shown an increasing openness to collaborating with districts to improve the appraisal of teachers (Johnson, Donaldson, Munger, Papay, & Qazilbash, 2009; Weingarten, 2010).
Another sign of hope is in the districts that are making progress in tightening the link between teacher evaluation and improved instruction and—potentially—achievement. The three districts I describe here have taken significant steps toward mitigating some of the current problems in teacher evaluation. They represent a new direction in evaluation that, if it spreads, can transform teaching and learning.
Cincinnati
In Ohio, Cincinnati's Teacher Evaluation System (TES) exerts an influence on instruction and, it appears, student achievement (Kane, Taylor, Tyler, & Wooten, 2010). The evaluation system grew out of the 1997 collective bargaining agreement between the Cincinnati Board of Education and the Cincinnati Federation of Teachers. With careful study and advice from experts in research and practice, Cincinnati developed a program for its 58 schools and approximately 2,200 teachers that has clear evaluation criteria and a structure that mitigates some of the common problems with teacher evaluation.
Using Charlotte Danielson's (2007) framework as a guide, Cincinnati has built its evaluation criteria on 16 standards that are arrayed within four domains: (1) Planning and Preparing for Student Learning, (2) Creating an Environment for Student Learning, (3) Teaching for Student Learning, and (4) Professionalism. The system devotes considerable time and resources to providing professional development on these standards.
Teachers in the district can apply for a three-year term as an evaluator or a consulting teacher. Teacher evaluators conduct three and administrators conduct one of the tenured teachers' four formal observations during the comprehensive evaluation cycle, which tenured teachers undergo every five years. Two of the observations are announced, and at least two are followed by post-observation conferences with evaluators.
New teachers and struggling veterans are evaluated under a different, but related, system. The district's Peer Assistance and Evaluation Program assigns consulting teachers to these teachers; these consulting teachers usually also serve as evaluators for other teachers, but they evaluate new or struggling teachers more frequently.
This system reduces the time problem many evaluators face by dividing evaluation responsibilities between administrators and teachers. Each full-time evaluator typically has a caseload of 18–25 teachers. Because they are focused on evaluating and assisting their peers, these individuals have the time to conduct high-quality evaluations and provide useful feedback to teachers. New consulting teachers and teacher evaluators receive 10–11 days of training before starting their work. Continuing evaluators receive five days of training each summer. Before their first term, all evaluators and consulting teachers must pass an evaluator certification test that requires them to assess instruction using the system's rubrics and demonstrate their reliability as raters. Over the course of the school year, consulting teachers and evaluators receive two hours of training every other week to review evaluation standards and calibrate scoring. After three years in their role, consulting teachers and teacher evaluators return to full-time teaching positions, which keeps their knowledge of teaching and learning current.
Northern U.S. Charter Management Organization
Some charter schools have also attempted to make teacher evaluation a more powerful tool for instructional improvement. One charter management organization in the northern United States, a successful network of 15 urban schools serving high percentages of low-income and minority children, has done so by deemphasizing formal summative evaluations and focusing instead on ongoing informal evaluation and feedback (Donaldson & Peske, 2010).
In this organization, teachers receive one-on-one and small-group coaching from administrators weekly or biweekly, as well as a midyear summative evaluation. The coaching is differentiated according to the teachers' needs and aimed at developing teachers' skills over time.
For the summative appraisal, evaluators and teachers complete the same six-page document. This appraisal form focuses on the organization's Aspects of Instruction, which covers such approaches as differentiation and checking for understanding.
Comments on the appraisal do not simply reflect a short period of formal observation, as those of some evaluation systems do. Instead, the document prompts both the teacher and evaluator to reflect on all the work the teacher has done so far that year. Thus, evaluators may draw on all their observations of the teacher—inside or outside the classroom, brief or sustained. This includes not only classroom instruction but also noninstructional contributions to teams and committees and to the school as a whole. Teachers reported spending three to five hours preparing the document and another 90–180 minutes debriefing with their evaluator.
The charter organization enables administrators to spend considerable time observing, evaluating, and coaching teachers by keeping the teacher to evaluator ratio quite low—approximately six teachers to one administrator. The organization has also strategically aligned personnel to handle certain administrative tasks so that principals can focus on instruction. An operations team handles facilities management, budgeting, certification, and ordering; a dean of students manages student behavior challenges; an intervention coordinator organizes schoolwide data and testing; and in some cases, an executive assistant coordinates activities with teachers and students and fills in where extra help is needed.
The organization lives by the principle that, as one teacher noted, "feedback is a gift." A key part of professional development focuses on training teachers and leaders to have difficult conversations, which sometimes occur during evaluation debriefs. One principal explained that her school has adopted explicit norms such as "staying on your side of the net and not stepping over and making claims on the other person" that help to depersonalize disagreements. Another principal said that in the process of hiring teachers, he deliberately gives them critical feedback on their demonstration lesson to see how they handle constructive criticism.
Evaluators receive training in how to deliver feedback in such a way that their suggestions will be implemented. They learn to give concrete and specific feedback that teachers can immediately respond to. Evaluation and coaching sessions deliberately focus on one or two major issues a teacher needs to work on and are anchored in student data, often the organization's benchmark assessments. This narrow focus helps teachers make changes.
Camden and Rockport
The Five Town CSD and Maine School Administrative District #28, under Patricia Hopkins's guidance, have also strengthened their evaluation systems. Hopkins notes, "I have seen a shift. People aren't just saying 'you're doing great.' They're posing questions and making recommendations to help inform teachers' efforts to improve instruction."
This shift has come about in large part as a result of the district's efforts to solve one typical problem of evaluation systems—lack of evaluator will. Early on, Hopkins decided to increase evaluators' accountability for completing high-quality assessments. First, she posted a calendar in her office showing the names and due dates of all teacher evaluations throughout the district. This calendar enabled her to keep track of and follow up with evaluators during the school year.
Administrators must conduct at least two observations each year for first- and second-year teachers and one every third year for teachers on a continuing contract. The evaluations themselves are based on both these observations and such factors as "promptness and accuracy of reports" and "evidence of professional growth." Administrators meet with teachers before and after the observations and evaluations.
Second, Hopkins and the assistant superintendent began to informally observe all first- and second-year teachers in the district. This practice of providing another set of eyes helped school-based administrators be more critical. In some cases, Hopkins said, the informal observation led to additional observations of teachers and more in-depth conversations with administrators.
Last, Hopkins has required principals to share their draft evaluation reports with assistant principals and vice versa before the post-observation conference with the teacher. This sharing has enabled administrators to clarify their expectations, maintain consistency with one another, and ensure that their commendations and recommendations for improvement are appropriate.
On the Road to Improvement
A key feature of all of these organizations is their commitment to improvement. Leaders in these schools acknowledge that their evaluation systems have room to grow. All the leaders noted that consistency of ratings, for example, could be improved. Overall, however, these schools have made progress on problems that have consistently plagued efforts to link evaluation and instructional improvement.
To varying degrees, they have also taken steps to integrate strengthened teacher evaluation systems with other key human capital efforts at the school and district level. Integrating evaluation with professional development or hiring, for example, helps ensure that the system as a whole functions more effectively; as a result, instruction improves substantially. It also make it more likely that improvements in teacher evaluation will endure. In fact, improving teacher evaluation without attending to other ways in which schools or districts affect teacher quality will likely have limited influence.
With shifts in the teaching force, the technology used in teaching, and the political and educational climate, a window of opportunity for improving teacher evaluation has opened. Some districts are already starting to move forward. The question is whether other districts will follow.
