Houston boasts one of the better urban public school systems in the United States, along with possibly the best charter school sector in the nation. Houston incubated the nationally successful KIPP charter schools, the regionally successful Harmony schools, and the less heralded—but highly effective—YES Prep public schools.
Founded in 1998 by Teach for America alumnus Chris Barbic, YES Prep has 13 campuses in Houston with 600 teachers serving grades 6–12. Ninety-nine percent of the 8,000 students at these schools are minorities and 83 percent are economically disadvantaged. YES Prep succeeds at its college-prep mission; an estimated 77 percent of its alumni have earned or are progressing toward a bachelor's degree—roughly seven times higher than the percentage of economically disadvantaged U.S. young people who pursue a degree (Maranto, 2013). Comparisons of student growth also rank YES Prep among the top charter organizations in the country (Woodworth & Raymond, 2013).
In part, YES Prep's success reflects its "hedgehog" growth strategy. While Harmony spread regionally and KIPP spread nationally, YES Prep stayed in Houston, growing slowly and fine-tuning systems to ensure coherence across campuses. Indeed, teachers refer to YES Prep's 13 campuses as "the district."
A High-Stakes System—That Teachers Like
One less-highlighted aspect of YES Prep's success, which I discovered serendipitously while doing teacher interviews there, relates to school morale—and seems counterintuitive. Teachers overwhelmingly support YES Prep's system for evaluating and developing teachers, although the stakes are high for teachers who receive poor evaluations. Ninety-three percent of YES Prep teachers I surveyed agreed with the statement "my principal or dean has helped improve my teaching" (57 percent strongly agreed). In contrast, of the nine teachers who had taught elsewhere, only one reported that administrators at her previous school improved her teaching. Teachers feel valued: 75 percent agreed that "leadership consults with teachers regarding matters that affect the school," 89 percent agreed that "I am treated as a valued employee" and 78 percent said, "I look forward to each working day at this school."
These teachers' favorable views of their evaluation and development system is surprising, considering the mixed record of teacher evaluations generally. Indeed W. James Popham (2013) playfully titled his useful guide Evaluating America's Teachers: Mission Possible? Support for this evaluation system is even more remarkable because teachers who receive negative ratings face consequences. At the end of each school year, some teachers who haven't scored well on their evaluations—up to 10 percent—aren't asked back. YES Prep ignores Popham's generally sensible advice to separate formative evaluations (aimed at improving teaching) from summative evaluations (aimed at rewarding or "deselecting" teachers).
Popham fears that the latter kind of evaluation taints the former, causing teachers to try to impress evaluators rather than try to improve teaching. It's a realistic fear, especially when systems involve higher salaries for teachers who receive better evaluations, as is true at YES. Indeed, teacher-evaluation systems and merit-pay plans have rarely improved teachers' performance. Recent evaluations of the impact of teacher merit pay on student learning found the effects to be positive, but modest (Barkowski, 2013; Ritter & Barnett, 2013).
YES Prep has overcome the many barriers to creating—and getting buy-in for—a teacher-evaluation plan using merit pay. Its approach has some applicability to traditional schools. Before examining YES Prep's plan, let's consider those barriers.
Why Merit Pay Fails
Teacher evaluation and its often ugly twin, merit pay, form a significant part of the Obama administration's education agenda. On the face of it, it seems like common sense that teachers will be more likely to improve if they're given serious—and at times critical—feedback. Ritter and Barnett (2013) argue that in the long term, merit pay will improve schools by focusing teacher efforts and attracting more talent to teaching. After all, a profession that pays everyone the same no matter they perform might not attract the most capable.
But stringent evaluations and merit pay often underperform for five reasons. First, merit-pay plans are often poorly designed. Effective plans should involve substantial sums of money; reward both group and individual performance to encourage collaboration; and use multiple, credible measures of teacher performance. Most don't deliver on these elements.
Second, even where a merit-pay system is well constructed, it depends on difficult, often subjective evaluations of teachers. Teacher evaluations must be transparent, involve ongoing and trustworthy communications between rater and teacher, and use a range of valid measures of teaching effectiveness, including growth in student scores on standardized tests, classroom observations by trained evaluators, and evidence like the quality of lesson plans (Popham, 2013; Ritter & Barnett, 2013; see also the November 2012 issue of Educational Leadership). Accurately evaluating teachers is a more resource-intensive endeavor than many schools can manage; little wonder many avoid it by rating nearly everyone outstanding!
Third, since the 1990s, schools have used sensible evaluation rubrics developed by consultants like Charlotte Danielson and Robert Marzano. Yet schools of education have been slow to train leaders to evaluate teachers, much less to use evaluations to guide personnel decisions. Leaders who want to reward or terminate teachers on the basis of performance thus lack an institutional infrastructure for support.
Fourth, as Charles Payne (2008) shows, innovations like merit pay seldom last long in poorly performing schools. Teachers and leaders wait the reforms out rather than changing behavior.
Finally, teachers won't respond to critical evaluations and attendant incentives unless they believe they have the efficacy to make changes and that their efforts can affect student learning. Unfortunately, as Payne's fieldwork (2008) suggests, in many high-poverty schools teachers simply don't believe their students can learn very much—-no matter how hard teachers work.
Why It Works at YES Prep
So how has YES Prep forged a high-stakes evaluation plan that truly improves instruction? Several elements are key to the system's effectiveness.
A Well-Crafted Rubric
YES Prep's evaluation form, the Instructional Excellence Rubric (IER), was fashioned over several years, field tested, and built within an existing culture of teacher development so it has credibility. The rubric applies four possible ratings (Unsatisfactory, Approaching Proficiency, Proficiency, and Mastery) to each of 17 specific indicators within two broad domains, (1) Classroom Culture and Management and (2) Instructional Planning and Delivery. Classroom Culture and Management is divided into five subdomains: Promoting Positive Climate, Sense of Urgency, Routines and Procedures, Student Behavior and Teamwork, and Physical Environment. Instructional Planning and Delivery has 12 subdomains, such as Content Delivery, Questioning Strategies, Providing Feedback, and Tracking toward Goals.
The rubric enables evaluators to accurately rate teachers' performance on each of these indicators. For example, an Unsatisfactory rating in Tracking Toward Goals indicates that a teacher "does not set goals for class, or sets goals that don't use appropriate data or resources to be sufficiently ambitious." In contrast, a Mastery/Exceptional Quality rating signifies that "goals are ambitious and result in significant student gain of knowledge or skills; Students can articulate goals and their importance to the class and their own intellectual development."
Each teacher is observed at least six times for evaluation throughout the year by his or her academic dean. A teacher's rating on each behavior on the rubric is determined by averaging the ratings that teacher earned on all observations (and less formal "walk-throughs") during the year. The final rating isn't purely a mathematical average; it takes trending into account.
Ratings on a third domain of the IER, Values and Responsibilities, aren't based on this rubric, but are determined by campus leadership teams, with a teacher's self-evaluation taken into account. YES Prep is now pilot testing the use of student test scores in teacher evaluations.
Personnel Decisions Grounded in Evaluation
On its own, this rubric is nothing extraordinary. Though somewhat less constructivist, it resembles rubrics developed by Marzano and Danielson. What is extraordinary is the degree to which the rubric guides personnel decisions, something rare in traditional public schools and even charters. Up to a tenth of teachers are terminated annually because their evaluation results haven't met expectations for a teacher at their experience level. Firings shouldn't be a surprise: Teachers receive regular feedback from those who observe them (weekly for new teachers); they know what their goals are and what they should do to get there.
All teachers fall into one of five levels on the school's personnel continuum: New to Blue (new to YES but with experience elsewhere), Novice, Developing, Practiced, and Advanced. Expectations of what a teacher at each level can do and the minimum scores a teacher can receive on observations are spelled out, and there are transparent criteria for moving up the levels. For instance, Novice teachers learning their craft generally move to Developing within a year. If a teacher's evaluation scores aren't high enough to advance in one year, he or she can stay at Novice level for two years, a reasonable period of time to improve. Teachers classed as Developing can stay on that level for four years before they must earn high enough marks to reach Practiced status, and so on. If there's no improvement, administrators ask that teacher to pursue other opportunities.
Evaluations and teacher levels determine pay. Novice teachers make $44,000 annually, comparable to salaries of new public school teachers in Houston. Advanced teachers' salaries begin at $66,000, roughly the salary of a 30-year veteran in the Houston school district. As Advanced teachers continue to serve more years, their salaries increase yearly. YES Prep currently has nine Advanced teachers.
Teacher Development
Most important, YES Prep has capable and trusted evaluators who also foster teachers' growth. These academic deans are highly effective teachers who are carefully vetted; they have strong instructional vision and can go into a classroom and realize what needs to happen to move the class forward.
Deans have cemented a reputation for helping teachers develop. More than 80 percent come from within YES Prep. Deans supervise about 15 teachers in their content area, which gives them credibility, and are frequently in classrooms, offering formal or informal feedback. Each teacher meets individually with his or her dean to be coached toward goals the teacher sets.
Deans are the main staff who observe and evaluate teachers, but YES administrators told me the most important thing deans do is build relationships with teachers. Because of these trusting relationships, deans can be mentors and evaluators.
School Qualities
Three qualities of YES Prep also allow this evaluation system to work.
- YES Prep is a young, growing organization—and a charter organization to boot. New organizations, with few rules and even fewer preexisting expectations, are easier to change.
- The school has had unusual stability of leadership: Founder Chris Barbic served as CEO for 12 years and much of the school's senior leadership has been at the school since its founding, or nearly so. Stability enables long-term thinking.
- YES Prep carefully recruits and hires its teachers, which may explain teachers' unusually high efficacy: 86 percent of them disagreed that "most of the children at this school are simply not capable of learning the material." The school hires most of its teachers from alternative programs like Teach for America (Maranto, 2013). Some evidence indicates that alternatively trained teachers may be more willing to take risks in terms of evaluations than traditionally certified teachers are (Bowen, Buck, Deck, Mills, & Shuls, 2013).
Should You Try This at Home?
YES Prep has built a strong evaluation plan that its teachers support not through remarkable evaluation techniques, but through keeping a stable mission, hiring and mentoring teachers who believe in that mission, and developing deans with the skill to evaluate and mentor teachers. This system wasn't imposed from outside the school's culture; rather, it reflects and reinforces that culture. So it's worth asking: Could the YES Prep system—or some parts of it—make teacher evaluation stronger without harming teacher morale in traditional schools?
The answer is a resounding maybe. The obstacles I've described make serious teacher evaluations problematic in traditional schools, where increasing pay for highly effective teachers and "deselecting" the ineffective is fraught with political peril. That said, public schools can—and should—adopt at least two of these practices: Schools can
- Hire for mission, as do YES Prep and charter organizations like KIPP (Shuls & Maranto, 2013). In the long term, such hiring might produce teaching staffs with greater efficacy and willingness to take risks, first steps in building an evaluation-friendly culture.
- Assign people other than the principal to observe, evaluate, and mentor teachers, as YES Prep's deans do. Principals typically lack the time to conduct thorough evaluations. Public secondary schools should select trusted department chairs with good teaching records, emphasizing the teacher mentoring mission of the position. At the elementary level, assistant principals—provided they mentor a limited number of teachers—might serve the same role.
Such reforms might be built on the district level, but probably only over a period of years, driven by long-term, capable leadership willing to take risks for reform. Innovative district leadership would also need to offer legal help to withstand challenges, which are likely when teachers are fired based on evaluations with teeth. On the state level, state education authorities could do more to offer training in evaluation.
These are difficult requirements. But in the matter of teacher evaluation and merit pay, there are no shortcuts. The evidence that YES Prep's teachers simultaneously experience stringent evaluation and good morale shows that schools can get there.ν
