Recovering from HSPS (Hyperactive Superficial Principal Syndrome): A Progress Report

In a previous article, Mr. Marshall shared with Kappan readers the innovative system he developed during his early years as a principal -- short classroom visits followed by informal feedback. But that system alone, he found, was not enough to bring about significant gains in student achievement. In the intervening years, he has learned that four complementary activities are necessary to achieve the primary goal of instructional leadership.

By Kim Marshall

SEVEN YEARS ago, I wrote an article for the Kappan that was featured on the cover. It carried a long-winded title: "How I Confronted HSPS (Hyperactive Superficial Principal Syndrome) and Began to Deal with the Heart of the Matter." In the piece, I described my unsuccessful struggle as a rookie principal to get into classrooms and talk substance with teachers. Then I shared my discovery of what seemed like a good solution: making five-minute visits to five classrooms a day and tracking down each teacher later on (within 24 hours if possible) for a follow-up conversation. The article did not generate much mail, but it has been widely read and discussed, and (perhaps not as a direct result) the idea of short classroom visits seems to have gained some acceptance.1

How have I handled teacher supervision since 1996? Did I keep doing five-minute visits? How did teachers react? And most important, did this system make a difference for student achievement?

An Update

The short answer is that I kept up my visits, averaging four a day (about 500 a year) for a total of eight years. Teachers were quite positive about this approach to supervision, and I firmly believe that it was an essential part of my instructional leadership. But classroom visits and feedback did not by themselves produce significant student achievement gains. I'll return to this point below.

First, some background. In my early years as a principal, I was constantly swamped by over-the-transom demands on my time -- a weeping girl with a splinter under her fingernail, a fight in the cafeteria, a teacher going through a personal crisis, a dog sneaking into the school, a parent cursing to the secretary about a bully on the bus, a jammed photocopier, a paraprofessional having a seizure, the laminating machine grabbing the end of my favorite tie, a call from the central office in support of that angry parent, a delivery from a trucker who refused to lift anything, and more. I aspired to be an instructional leader, but I simply didn't have the time to do teacher supervision the way I'd been taught it was supposed to be -- the canonical preconference, whole-class observation, and postconference debriefing. I also realized that there was no way that I could be the staff developer for the school, as I had naively hoped I would be.

Chatting with other principals and perusing reality-based professional literature, I realized that I was not alone. The dirty little secret of American schools is that principals rarely get into classrooms. Powerful, almost inexorable forces conspire to keep school administrators from a meaningful instructional role. The result is Hyperactive Superficial Principal Syndrome (HSPS), and it looks like this:

* the principal is trapped in the office dealing with one crisis after another;

* if the principal escapes, he or she wanders around without a clear agenda and misses a lot;

* each day is so chopped up by interruptions that it's very hard to focus on deeper stuff;

* teachers are rarely observed in a thoughtful way and almost never get feedback;

* evaluation visits happen only when they are absolutely required, which is usually once every year or two;

* evaluators often see "glamorized" lessons that aren't very representative of daily practice;

* evaluations are sometimes based on rumor, gossip, and innuendo and are not very deep;

* there are few authentic teaching/learning conversations between teachers and principals;

* teachers are mostly on their own and get used to working in isolation; and

* starved for feedback, many teachers stagnate, and mediocrity flourishes.

The result of this pattern is that certain "natural" forces in schools tend to run rampant. In short, the rich get richer, and the poor get poorer. Students who enter school with disadvantages learn very little and fall further and further behind. And the gap between the haves and the have-nots gets wider each year.

An instructional leader -- that rare gem -- successfully pushes back against these forces and narrows the achievement gap. It's not easy. It takes guts and self-discipline. It takes incredible time management and heroic leadership. But it can be done. A regular regimen of short classroom visits is a good place to start.

The reaction I often get when I describe my quick observations is, "What can you see in only five minutes?" Let me put this in perspective. The length of time a principal needs to spend in a classroom depends on the purpose of the visit. If the principal is checking up on a substitute teacher, five seconds is long enough. (He's up and teaching, the kids are in their seats, and nobody is fighting, so all is well.) If the principal's purpose is to "show the flag," 30 seconds is plenty. (I'm the principal, I'm alive and well, and I care about all of you.) If the purpose is a formal evaluation visit, 40 minutes is probably required by the contract. (This depends on the school district, and I'm happy to report that Boston recently did away with this requirement, allowing multiple short observations to fill the bill.) And if the purpose of the principal's visit is to build a case for dismissing a teacher, it's going to take multiple 40-minute visits, each followed up with a diagnosis and prescription and a chance for the teacher to improve between visits.

But if the principal wants to get a general sense of how a teacher is doing and then have a substantive follow-up conversation about a particular teaching moment, five minutes is plenty. True, it's a mere sliver of a teacher's day. But five minutes in a classroom is a long time. When I speak about teacher supervision to a graduate class or an audience of principals, I often play a videotape of a teacher in action and turn it off after five minutes. People are astonished by how much longer it seemed and how much happened in that time. The challenge for the observer, they agree, is deciding which of seven or eight possible "teaching points" to pursue in the follow-up talk. My experience has been that, if I stay in a classroom for 10, 15, or 20 minutes, there are diminishing returns. For principals who are constantly being pulled in a hundred different directions, longer visits are a luxury that they simply can't afford, and long visits are not the best use of their time.

Principals who make short, frequent visits can see a lot. But do they see enough? This begs the question of what principals need to know about teaching and learning. Here's a list of questions that school leaders should be able to answer:

* Are teachers on track with the curriculum?

* Are the students learning?

* Are teachers "happy campers" in terms of their jobs and their lives?

* Do some teachers deserve special praise?

* Do some teachers need redirection, emergency support, or a negative evaluation?

A principal cannot possibly answer these questions without spending quality time in classrooms and having substantive follow-up conversations with teachers. Just asking a teacher how things are going isn't going to get below the surface. But spending hours in classrooms every day may be overkill -- and may pull the principal away from other kinds of leadership that are essential to improving student achievement. I believe that a regular cycle of five-minute classroom visits with a follow-up conversation after each one is the most efficient way for a principal to monitor classrooms and find the answers to those key questions.

There is another reason for keeping classroom visits to only five minutes: it's the only way to visit frequently enough to see the big picture. If visits are lengthy, the principal can't fit as many into each day and won't see teachers often enough to have a sense of each teacher's overall reality and the reality across all classrooms in the building. Keeping up a pace of five classrooms a day, I was able to visit all 45 teachers in two weeks and then start another cycle. Seeing each teacher once every two weeks is far from ideal. Indeed, it amounts to just half a percent of total teaching time. But because it is a representative and random sampling, it gives an excellent sense of what is going on.

Observations on Observations

The biggest challenge posed by classroom visits is not length -- it's depth. Here are some reflections on this and other aspects of classroom supervision from my 15 years as a principal.

* What I found hardest was to slow down, get other worries out of my head, and smell the roses. This meant being sharp and fresh -- yet another reason for getting enough sleep, exercise, and down time. When I was "in the zone," I could see a huge amount in just a few minutes. When I was tired, stressed, and distracted, I could "observe" for a whole period and see very little.

* Getting into four or five classrooms a day was always a struggle. Like a recovering addict, I sometimes lapsed back into HSPS and went for several days without visiting a single classroom. It took a lot of self-discipline to keep on track.

* My superiors never bought into the system. Not once did my boss ask me how many classrooms I visited each day, how long I stayed, how I gave feedback to teachers, how I kept track of my visits, how many cycles I aimed to complete in a year, or how my visits were factored into formal evaluations. These are all good questions to ask any principal. With no encouragement or prodding from outside, it was completely up to me how often I visited classrooms. The only thing that kept me going was faith that the visits were good for teaching and learning.

* I found that having a numerical target (five a day) was the key to getting into classrooms and pushing back (at least for those 25 minutes) the incessant tug of other duties and demands. Without a measurable goal in mind, a day would zoom by without my having observed a single teacher.

* In terms of time management, I found it most efficient to fit in my brief visits on the way to and from other errands and expeditions around the school. Sometimes I was successful in blocking out a whole period for classroom visits, but that amount of time rarely went by without my cell phone ringing or something else coming up. Still, with dogged persistence, it was usually possible to squeeze five visits into the nooks and crannies of each day.

* Over time, I got better at seeing what was really going on in each classroom and sharing one interesting insight with the teacher when we talked later on. Some of the follow-up conversations were short and awkward, but many went into great depth about a particular teaching moment, what teachers were trying to accomplish with the lesson, how their kids were doing, and what their dreams and fears were.

* I found that verbal feedback worked best. When I occasionally gave feedback in a written note or an e-mail, the teacher rarely responded, so there was no dialogue. I also found that I was much more guarded about putting negative feedback in writing because it was more permanent (and more threatening) and because I couldn't judge the teacher's mood and ability to absorb a critical comment. Written feedback also deprived me of the rich and substantive conversations that often grew out of face-to-face feedback talks. So I stuck almost entirely to verbal follow-ups, usually stand-up conversations in such informal settings as the teacher's classroom during a free period, in a corridor, by the copy machine, or while walking to our cars after school. These talks lasted an average of three to five minutes.

* For many teachers, these talks were the only adult conversation they had about their work; my feedback helped to some degree to relieve the loneliness and isolation that are an inevitable part of teaching. Regular feedback also gave teachers a sense of their part in the broader learning community and contributed to their sense of efficacy, and the teachers had a chance to test their ideas, to push back against mine, and to spread their professional wings a little.

* The visits and follow-up talks also brought an important psychic payoff to me: when I managed to get four or five solid visits and feedback talks into a day, I felt like an instructional leader. I saw several classes of students in a positive setting (which was not the case when I was dealing with discipline, stopping students from running in the halls, and supervising the cafeteria), and that alone made me feel like an educator rather than a cop.

* The feedback conversations played an important ethical role in this system: they gave teachers a chance to clue me in on the broader context of a particular teaching moment, to correct me if I misheard or got the wrong idea, and to push back if they disagreed with a criticism. When a principal does not follow up after a substantive classroom visit, the teacher is left guessing and can get very anxious and frustrated. Call it supervisus interruptus.

* My visits also kept me in touch with the curriculum and made me a more effective participant in grade-level and school-site council meetings. Being able to give concrete examples of teaching and learning I had seen in classrooms made me a much more credible player in these crucial discussions.

* I found that, every year, there were a few classrooms that I avoided until the very end of a cycle. Only my self-imposed goal of finishing one cycle before starting the next forced me to make my way into these unwelcoming, mediocre rooms. After my reluctant visits, I tended to avoid these teachers, sometimes procrastinating for days before giving them my very mixed feedback.

* Giving negative feedback never stopped being difficult. I tried to follow a friend's advice: take a deep breath and start talking. But I still pulled my punches with some hypersensitive teachers. In retrospect, perhaps it was smart to avoid provoking a defensive and unproductive reaction.

* When I gave critical feedback, I sometimes confessed that I had made the same mistake when I was a teacher. When I gave advice, I tried to make it easy for the teacher to put into practice. This helped teachers accept and act on my comments.

* I found that frequent visits were the key to being able to deliver honest criticism to teachers without sending them into a tailspin. The hardest thing for a teacher to handle is getting negative feedback when the administrator hasn't visited in three months and hasn't seen hundreds of successful teaching moments. Sensing this potential upset, some fair-minded administrators who have been guilty of not getting into classrooms tend to shy away from critical comments, sugarcoating their criticisms or reaching for something positive to say. But if the principal is making a dozen visits a year and nine of them are followed by genuinely positive comments, it's a lot easier for a teacher to hear criticism after the other three.

* I was rarely successful at having a "theme" in a series of classroom visits -- visiting all the fourth-grade teachers or all the bilingual classrooms in a day or looking at "wait time" or computer use for a whole 45-classroom cycle. Doing this would have sharpened my observations and given me a better perspective on particular aspects of the school. It's definitely something I recommend.

* After a few years, the staff got to like the system so much that virtually all teachers were comfortable having me use my brief observations to write their official performance evaluations. We got away from the dog-and-pony shows of contrived, unrepresentative lessons designed solely for my benefit.

* I believe that the time I took observing classrooms and giving detailed, clinical feedback created a more positive staff culture, which in turn made our school more attractive for good teachers looking for jobs. Over the years, we attracted a number of very talented new staff members.

* In sum, I remain a strong proponent of principals' getting into classrooms in a systematic way and giving teachers honest, detailed, face-to-face feedback soon afterward. It is vital to running a good school and being an instructional leader.

BUT over the years, our students' achievement, while improved, did not reach the level we wanted. Too many of our students were still in the failing category on the state's 4-3-2-1 scale, and far too few were in the proficient and advanced levels. Was I just spinning my wheels when I put so much time and energy into my supervisory visits?

I have come to believe that there are four complementary interventions that need to be in place for a principal to get real traction from classroom visits: 1) clear grade-by-grade curriculum proficiency targets, 2) teacher teams that plan curriculum units with an end in sight, 3) teams engaging in Japanese-style lesson study, and 4) a "power learning cycle" in which teachers use student work and data to improve teaching and learning. I was not successful in getting all four firmly in place, but I believe that, if I had, our student achievement would have been far more impressive. I'd like to explore each one in more detail.

Curriculum Clarity

When a principal visits a classroom, one of the most important things to looks for is whether the teacher is on target with the curriculum. Of course, to answer the question means that the principal must know what exactly the curriculum is! If principals don't have a clear sense of what second-graders are supposed to learn and what good student work should look like in, say, December, it's awfully hard to supervise teachers effectively, and a lot of time can be wasted. The tendency in American schools (in contrast to those in most other countries) has been for the curriculum to be quite vague and for teachers to have a great deal of latitude to create their own curriculum behind closed classroom doors. It's very hard for a principal's classroom visits to address this kind of anarchy piecemeal, a single teacher at a time. Teacher supervision can't be efficient and effective until curriculum expectations are clear and widely accepted within the school.

Fortunately, the standards and high-stakes tests adopted by most states in the last few years have made curriculum anarchy untenable and created a stinging mandate for school districts to spell out exactly what students must know and be able to do at each grade level. Many districts are wrestling with the challenge of creating new curriculum documents, and it's not easy. Curriculum writers (and especially curriculum committees) tend to fall into one of four traps. First, they often churn out impressive but unwieldy documents that contain far more objectives than can possibly be covered in a year, putting stressful demands on teachers; as a result, some schools still have a different curriculum in each classroom because teachers make idiosyncratic decisions about what is most important and what they're most comfortable with. Second, curriculum documents from the central office are often vaguely worded, poorly written, and filled with jargon, making it difficult for teachers to translate them into action. Third, curriculum writers can rarely resist the temptation to get into the "how to" as well as the "what" of curriculum, spinning out elaborate units and lesson plans (which teachers rightly regard as advisory) and muddying the waters as to what students must know. And finally, curriculum writers find it hard to let go of previous documents in which they invested time and energy before the arrival of new state standards, and sometimes they falsely assure their superintendents that a cosmetic revision of the old curriculum is perfectly aligned with the new external standards. (Few superintendents know enough about the nuts and bolts of curriculum to avoid being bamboozled by presentations of this kind.)

What teachers and principals need are crystal-clear descriptions of what student proficiency looks like at each grade, reduced to an essential core, tightly aligned with state standards and assessments, written in jargon-free prose that is understandable to any parent, and accompanied by scoring guides and exemplars of actual student work at specific levels of performance. With documents such as these in hand, teachers can be part of a team effort and should be blessed each September with students who have a better and better grasp of the building blocks for success at the next grade level. In a school that has this caliber of grade-by-grade curriculum targets, the principal can visit classrooms with a much more discerning eye and be able to see immediately whether teachers are "on message" with the grade-level goals for that year. This should be evident even in a brief classroom visit. Principals shouldn't have to waste their time dealing with such curriculum train wrecks as the rain forest unit being taught in second, third, and fourth grades. They should be able to focus on how well the required curriculum is being taught in each classroom -- and whether students are learning what's being taught.

Unit Planning and Student Products

During classroom visits, a principal, no matter how sharp-eyed and perceptive, will not pick up certain macro elements of instruction. It is often hard for an observer to get a sense of the bigger picture of the curriculum unit that the teacher is pursuing (sometimes because it doesn't exist!). Looking at lesson plans or talking to the teacher afterward may not add much value since lesson plans are often well-intended fictions and teachers aren't always articulate about their long-range plans. The danger here is that the principal will be seeing a lot of ad hoc teaching and will get bogged down (and overwhelmed) trying to elicit more purposeful teaching. In addition, this is not the best use of the principal's time. Well-planned curriculum units should already be in place.

This is why it's essential for teacher teams to be given clear responsibility -- and a well-thought-out protocol -- for designing thoughtful curriculum units aimed toward a specific goal. Before beginning a unit on, say, Colonial America, the third-grade teachers should have articulated the big ideas and essential questions that students must master by the end of the unit, and all of these must be aligned with the state standards and high-stakes tests. The teachers should have written the unit tests and planned student projects and performance tasks up front, and they should have a good strategy for teaching the material in the most compelling, involving, and effective way possible.2

In schools that have bought into systematic unit design and trained staff effectively, classroom teaching is much more focused, and a visiting principal can concentrate on the heart of the matter: the quality of instruction and the students' reactions to what is presented. The "essential questions" of the unit should be up on the wall, and any student, if asked, should be able to articulate what he or she is trying to learn. As part of good unit planning, some schools develop standard ways for students to store their work (e.g., a reading log, a reading response journal, a writing portfolio, a math folder). When these routines are in place, the principal can ask to see a student's work and know what to look for. Good unit planning and schoolwide routines governing student work make supervisory visits far more productive. Unit planning meetings should be run by teachers, but the principal should be a regular visitor, partly for quality control, but also for the pleasure of watching substantive, collegial meetings that address the heart of the matter -- teaching and learning.

Lesson Study

One of the downsides of brief supervisory visits is that teachers may think that the principal is not interested in how the whole lesson unfolds. This could lead teachers to be less thoughtful about preparing and teaching coherent lessons and lead them to focus instead on the "razzle dazzle" of showmanship. This would be most unfortunate. The art of presenting material and involving students in a well-crafted 45- or 90-minute lesson is at the heart of good teaching.

I believe that, most of the time, principals cannot afford to watch whole lessons. If they try to, the pace of their staff supervision will slow to just one or two teachers a day, and they won't get a representative picture of each teacher or the faculty as a whole. I also believe that the principal is necessarily a generalist and should not get bogged down looking at every lesson plan and commenting on the details of each lesson.

But teams of teachers can and should take on this challenge. Japanese schools have developed an excellent protocol in which teachers "polish the stone." That is, they develop, pilot, observe, and perfect individual lessons to address specific student learning needs. James Stigler has suggested ways to adapt Japanese ideas to American schools by using videotaped lessons as discussion tools for teams of teachers.3

High on any principal's list of staff development challenges is training teams of teachers to implement lesson study and providing time and stipends to make it a regular feature of professional meeting time. In schools where teachers craft effective lessons and evaluate their effectiveness, the quality of instruction will improve by leaps and bounds. In addition, teachers' sense of efficacy and professionalism -- the deep kind of morale -- will benefit from this kind of nitty-gritty, solution-oriented work on lesson design.

Principals can't be involved in every lesson study meeting, but they should drop in often enough that they can become familiar with teachers' developing sense of what works in the classroom and what constitutes a good lesson. With this shared culture, principals will see more and be able to comment more thoughtfully when they make classroom visits.

Using Data to Improve Teaching and Learning

One thing that's very hard for a principal to see in a classroom visit, however long, is whether the whole class is really learning what's being taught. The visiting principal can observe the lesson, question a few students, and get an intuitive sense of whether the kids "get it." But it's difficult to tell if the whole class understands and impossible to tell how student performance breaks down by gender, race, and income. Actually, very few teachers know the answers to such questions. I remember a tiresome pattern when I was a sixth-grade teacher in the 1970s: "Does everyone understand?" "Yes, Mr. Marshall." "Are there any questions?" "No, Mr. Marshall." And then on Friday most of the students did horribly on the test. Far too many of us move on without checking adequately for understanding. Grant Wiggins calls it the educator's egocentric fallacy: "I taught it, therefore they learned it."

When I became a principal, one of my weakest points was working with teachers to look systematically at student learning. I was constantly frustrated that I could not get a handle on student achievement in classroom visits. But classroom observation is not the best way to find out precisely what students are learning. I could have spent a great deal more time in supervising teachers and not had much more insight. We needed to have a much deeper process in place.

In a recent article, Michael Dubrovich, an elementary principal in Colorado, described how he has the teachers in his school submit students' scores on math unit tests (every three weeks), on reading unit tests (every six weeks), and on periodic writing prompts.4 Using spreadsheets, he is able to see the distribution of grades, talk to teachers about the patterns he sees, and supervise teachers on how well they are using data on student learning to fine-tune their instruction and to deliver extra help to struggling students. He assures teachers that they will not be evaluated on how much their students learn -- only on how well they use the data.

My first reaction to the article was enthusiastic; this is what I should have been doing all those years to keep tabs on results. But when I thought about Dubrovich's system some more, I wondered whether I could -- or should -- inspect every unit test and micromanage teachers' productivity to such a degree. It's not that we shouldn't be looking at results. Nothing is more important. But teachers should be responsible for using formative assessment data to improve instruction. Not only is there far too much data for a principal to manage effectively, but research from the business world tells us that Theory Y management (bottom-up, big carrot) works better than Theory X (top-down, big stick). One of the central tenets of TQM (Total Quality Management) is that "inspection" should take place during the production process and should be conducted by those who actually do the work -- rather than at end of the line by the boss.

The ideal scenario is for groups of teachers (the kindergarten team or the math department) to decide on their unit goals, write common assessments, set a standard for proficiency, and then, when the unit is finished, score each set of papers and sit together to take an honest, collegial look at the results, asking themselves such questions as, How many students are proficient and above? Which students failed and need to stay after school for extra help? Which teaching strategies are working and which are not? Why did the students in Room 202 do so much better than those in all the other rooms? Why do girls seem to be doing poorly across the board on this unit? What can we do differently on the next unit? For the next year? How are we doing at narrowing the achievement gap? This kind of meeting can be an integral part of a "power learning cycle" that includes aligned curriculum goals, results-oriented unit plans, and excellent teaching. Periodic assessment can drive a process of continuous improvement.5

Data-focused meetings won't happen on their own, of course. They don't come naturally to American teachers, and they run counter to the culture of most of our schools. It's much more common for teachers to ply their trade in isolated classrooms and feel under such pressure to cover the curriculum that they rarely spend time looking systematically at results. The principal's leadership is essential to launching results-focused meetings: training needs to be scheduled, teachers need to be given time, and everyone needs continuing encouragement and monitoring. The single most important thing a principal can do is to block out in August the basic assessment schedule for the year (with dipsticks at least every nine weeks) and get dates for meetings to share data on everyone's calendar. The goal of the effort is to create a low-stakes, mutually supportive environment in team meetings so that teachers can have an honest and open dialogue about student work and feel that it's all right to make mistakes. This is a wonderful way to open classroom doors and to help end teachers' isolation from one another. Meetings such as these (and also the unit planning and lesson study meetings described above) provide excellent staff development and should be counted as such in a school's professional development plan.

The principal won't be able to sit in on every meeting, but that's fine. Real ownership needs to reside with each teacher team. With results-oriented meetings in place and teachers constantly monitoring and using student learning data, a principal can focus on pedagogy and the dynamics of student learning during classroom visits. And the principal will also have a deeper perspective on what's going on in classrooms. (Sometimes a teacher who is not very flashy can produce excellent results.)

Richard DuFour, one of the leading advocates of data-driven teaching, argues that principals need to shift their emphasis from teaching to learning, "from helping individual teachers improve instruction to helping teams of teachers ensure that students achieve the intended outcomes of their schooling."6 I heartily agree with this shift of emphasis, but I don't think it's an either/or proposition. As principals focus more on results, they should not retreat from daily supervision of classrooms to the minimum number of evaluation visits required under the contract. They should strive to have the best of both worlds by adopting an efficient regimen of brief classroom visits every day -- always with prompt and thoughtful feedback to teachers -- and by periodically asking teachers and teams for evidence of their students' progress.

Mike Schmoker has suggested that all educators worth their salt should be able to answer two basic questions at key points in the year: What percentage of students are proficient or above? In which specific areas are students having difficulty?7 A teacher-led process of periodically looking at data on student learning is the engine of improvement, and the principal plays a key role by leading and scheduling the process and then by regularly inspecting for classroom quality and student learning results.

Putting It All Together

Every principal knows that the real work of a school is done by its teachers. But when teachers do their own thing without coordination and a sense of purpose, they are like an orchestra without a conductor. The challenge for principals is that most of the time their work doesn't put them in the spotlight, visibly directing the efforts of the school. Most of the principal's instructional role takes place behind the scenes -- hiring good teachers, making sure curriculum is aligned and effective, cajoling independent-minded teachers who are used to doing their own thing, bringing in good training, scheduling and orchestrating team meetings, and analyzing data.

Applause, recognition, fame, glory? Not too often -- which is tough for people with healthy egos. Deeply embedded in most principals' heads is the traditional paradigm of leadership. In the same way that teachers may feel self-conscious and inadequate if a visitor walks into their classroom and they're not "really teaching" (i.e., holding forth at the blackboard, all eyes on them), so principals can feel that they're not doing their job if they're not acting like a traditional leader. This feeling is yet another reason principals gravitate to the noninstructional realm, where they can raise their voices and command respect (sometimes with a bullhorn).

The reality is that principals can't inspire every child, observe every classroom, scrutinize every lesson plan, plan every unit, look through every student's portfolio, analyze the results of every test, lead every training workshop, and chair every team meeting. Given the impossible number of academic challenges and the even more overwhelming number of operational demands, principals have to empower teachers to do this work, and they have to be very strategic in the way they spend their time as building leaders -- otherwise, they will fall victim to HSPS and preside over a fragmented staff, abysmal student performance, and an ever-widening achievement gap.

So what does effective instructional leadership look like? The most astute principals make sure that crystal-clear, manageable, grade-by-grade proficiency targets are in place before the beginning of the year (hopefully crafted by the central office) and then provide strong leadership for regular team meetings in which teachers:

* plan each teaching unit with the end in sight,

* work together to craft highly effective lessons that address key learning challenges, and

* use student work and assessment data to fine-tune teaching and learning and get extra help for students having difficulty.

The principal is not present for much of this work, but monitors it by 1) sitting in on occasional meetings, watching model lessons, analyzing the data from sample units, and talking to randomly chosen students about their work; and 2) making daily classroom visits with candid feedback to every teacher. All of this is much more likely to happen if people have a regular routine. With regard to the principal's classroom visits, a numerical target for each day is essential to keeping up the pace and getting into all classrooms on a regular basis. With regard to the meetings of teacher teams, weekly meetings might rotate from one devoted to unit design, to one on lesson study, to one on analyzing student work and data, to one on logistics and student discipline issues.

In my 1996 article, I wrote about a "force field" that sometimes kept me out of classrooms. It came from my own sense of inadequacy and from some teachers' resistance to being judged, and it slowly dissipated over the years as I became a more confident observer and as teachers came to trust that I was really there to help and support them. This force field exists in most schools; along with Hyperactive Superficial Principal Syndrome, it prevents principals from getting into classrooms and having authentic conversations with teachers that might improve student learning. Principals need to push through the force field and be regular visitors to all the classrooms in their buildings, observing thoughtfully and giving teachers perceptive and helpful feedback.

Principals are much more likely to get into classrooms -- and their feedback will have much more impact -- if they have been skillful at getting teachers fully invested in clear curriculum goals, unit planning, lesson study, and the use of student data. This combination within a school should be synergistic and yield much higher student achievement than any of the elements could produce alone. And this will make all the hard work worthwhile.

1. Margery Ginsberg and Damon Murphy, "How Walkthroughs Open Doors," Educational Leadership, May 2002, pp. 34-36; and Nancy Protheroe, "Improving Instruction Through Teacher Observation," Principal, September/October 2002, pp. 48-51.

2. An example of a thoughtful and well-developed approach to planning units can be found in the "Understanding by Design" materials, developed by Grant Wiggins and Jay McTighe, which are available through the Association for Supervision and Curriculum Development (www.ascd.org).

3. James Stigler and James Hiebert, The Teaching Gap (New York: Free Press, 1999). See also Clea Fernandez and Sonal Chokshi, "A Practical Guide to Translating Lesson Study for a U.S. Setting," Phi Delta Kappan, October 2002, pp. 128-34.

4. Michael Dubrovich, "Student Achievement Data: Holding Teachers Accountable," Principal, March 2002, pp. 30-34.

5. Mike Schmoker, Educational Leadership, May 2002, pp. 12-15.

6. DuFour, p. 13.

7. Mike Schmoker, "First Things First: Demystifying Data Analysis," Educational Leadership, February 2003, pp. 22-24.

KIM MARSHALL was principal of Mather Elementary School, Boston, Mass., from 1987 to 2002. He is now a writer and consultant who is working to support effective leadership in urban schools (kim.marshall8@verizon.net).