The missing piece of education reform

Leadership by superintendents and principals is key

TWENTY-FIVE YEARS ago a broad coalition of legislators, business people, education experts, and state officials put together and passed a wide-reaching education reform law. That law reflects a set of shared beliefs—basically, that a combination of increased funding, state testing tied to graduation requirements, new state curriculum frameworks, charter schools, and increased authority for superintendents and principals would lead to better schools. I was part of that coalition, but later came to believe that the law omitted a critical element—working with principals and teachers to help them improve their craft – and helped start a non-profit (the Bay State Reading Institute) to work with elementary and middle schools in this way. My experience at the Bay State Reading Institute over the past 11 years with several dozen schools convinces me that current state policies—based on that unchanged policy consensus—are unlikely to lead to further improvement in our schools.

Results from the National Assessment of Education Progress (the so-called “nation’s report card”) show that in the years immediately after 1993 there was a measurable gain in Massachusetts students’ performance. But the results also show that scores have leveled off in the last few years and that there is still a sizable gap between the achievement of low-income students and their wealthier counterparts. Looking at international test comparisons, the Massachusetts Business Alliance for Education’s 2014 Brightlines report concludes that while our scores are higher than other states, Massachusetts students are falling behind the international leaders. The highest performing Massachusetts students are well behind those in the highest performing countries.

The Bay State Reading Institute worked last year with 50 Massachusetts elementary and middle schools (almost entirely in low-income communities) to improve student achievement. Of the 33 schools with us for more than a year or two, 25 have made impressive gains while the results at eight others are disappointing. Comparing the gainers with the disappointments leads me to four major conclusions:

The key factor limiting school improvement is not teachers, but leadership from the principal and the superintendent. Of course, improved schools require improved teaching. But in a school where the principal is committed to change, knows the pedagogy she (or he) is looking for, is willing to hold difficult conversations with teachers when needed, and has the backing of her superintendent, our experience is that the principal can always move her teachers. In each of the eight disappointing schools the problem was a lack of principal commitment, an inability or unwillingness by the principal to make specific pedagogical requests of teachers, or central office officials undermining the principal’s authority.

This finding about superintendents and principals has enormous implications for state policy. In its effort to improve schools, the Department of Elementary and Secondary Education should focus on superintendents (and school committees), since they hire principals and give them their marching orders. The department should not be looking at programs that directly influence teachers, since good principals can handle that far better than a state agency that is necessarily focused on enforcement. This is not a criticism of the Department of Elementary and Second Education staff, but instead reflects the agency’s institutional enforcement role; it’s hard to change human behavior if the person being helped is not at ease with the potential helper.


Given the right kind of help, the great majority of principals and teachers are capable of running great schools. This begins with dramatically higher expectations for student performance; after a year or two, many teachers tell us that their students are now performing at levels the teachers never before thought possible. It also means major changes in how teachers teach and principals lead, starting with having the class spend most of the day in small-group instruction—the teacher working with a small group of students (at just their instructional level) while other students collaborate on high-level center activities, such as group discussion of text (including making predictions, asking questions, looking for evidence, and picking out the main ideas), separating fact from opinion (1st grade!); and debating topics like genetically modified food (starting in grade 3!).

Contrary to popular perception, great teachers are made, not born. Investing in teachers pays! People choose teaching because they care about kids; teachers will try out new approaches if they feel listened to and have evidence that these approaches work. Principals who understand and respect these truths can move their teachers. The Bay State Reading Institute asks principals in its partner schools to discuss individual student data with teachers, including what they plan to do help students who fall behind and what help they might need. In the most successful schools, the principal knows the pedagogy she’s looking for, spends an hour or two in classes most days, and makes very specific requests to teachers as to improvements she’d like to see. As part of this process, teachers begin discussing with each other all of the students in their grade level and begin working together to make sure every child (not just strugglers but also gifted students) reaches his or her potential. Once they see that the focus is on helping students (and not punishing teachers), teachers are comfortable with these discussions, and we have not had union resistance.

School districts need additional funding to make the necessary investment. When budgets are tight, investment in better teaching—be it purchasing new textbooks, hiring instructional coaches, making time for teachers to plan together, or bringing in outside partners—is the first area cut. Few districts—particularly high-poverty districts—have the funds they need to make these investments. Sadly, though, additional state funds that districts could use only for investment (preferably, only for investment and not operations) are not on anyone’s priority list.

Current state education policies do not reflect these lessons and are unlikely to succeed. As any successful high-tech entrepreneur understands, you motivate an educated workforce by setting high goals and then delegating authority, listening to your employees, making it safe to take risks, being generous with praise, encouraging collaboration, and using data in a constructive way to know what works and what doesn’t. State policies applying these principles to superintendents would mean avoiding micro-management and one-size-fits-all regulations from the state. Except in a handful of particularly badly managed districts where the state should and does appoint receivers, the relatively small staff at the Department of Elementary and Secondary Education is unlikely to make better decisions for each particular district than the principals and superintendents currently running those schools.

State policy would be far more effective if it took the advice of the authors of the Brightlines report—“You can’t mandate excellence; you can only inspire it!” The state could inspire excellence by creating a competitive state grant program for districts that offered significant, long-term funding to superintendents who have exciting ideas for investing in their teachers and their schools. My suggestion would be $250,000 per school, with a modest local contribution. This would be one-time money that the superintendent would have four years to spend and, once awarded, could not be taken away by the state or diverted by the school committee. Over the course of a decade, $50 million a year of state funding (just 1 percent of the $5 billion in state funds spent annually on K-12 education) would be sufficient to offer this one-time transformational opportunity to every school in the Commonwealth.

Another way to motivate superintendents would be to waive state mandates for districts that offer compelling plans of their own. Mandates typically link state funding to specific policies; for example, a state literacy program last year required that, as a condition of receiving funds, districts use specific literacy materials, regardless of whether the materials are consistent with the literacy program the district has adopted or whether it was achieving good results.

I’d offer to waive just about any state mandate, as long as the superintendent has a compelling idea for change, has indicated specific educational goals such a waiver would advance, and is willing to submit data to measure progress. This would put the emphasis on the district’s success in improving student performance (and the superintendent’s leadership in this area), rather than on micro-managing how a superintendent runs her district.

The combination of long-term funding dedicated to investment in better teaching and the opportunity to replace rigid state mandates with locally designed programs to improve teaching will be a powerful motivator to the state’s best superintendents.

For this to be successful, we’d need a rigorous but open-minded assessment of the improvement plans superintendents put forward. It could be modeled on the high-quality evaluation process the Department of Elementary and Secondary Education built for its Priority Partner program, which includes well-respected outside reviewers.

Keeping in mind the critical role of district leadership, let’s take a look at the major education improvement policies currently in place:

  • State education officials have an excellent program to mentor new superintendents. The program has engaged some of the most respected retired superintendents in the state, who serve as coaches for new superintendents. The program should continue and be supported.
  • Massachusetts should retain a modest amount of statewide testing to make sure we catch those schools that fall hopelessly behind, but the amount of testing we do now is way out of proportion to the benefit. I’d test as little as possible in grades 3,5,6, and 7, and keep no more than four hours of testing in grades 4, 8, and 10. Through 2015, schools spread testing over two whole months; during those months, when tests were being conducted, teachers who normally were assigned to provide extra help to struggling students stopped those services so they could be test monitors. Test results also aren’t available until the following fall—too late to help teachers improve instruction for the students in front of them.

As currently constructed, the state reports on standardized test results tell us very little about the quality of teaching in any school. They tell us that students in, for example, Wellesley and Lexington have higher scores than students in Boston or Springfield. That’s not surprising given that, on average, students entering kindergarten in high-poverty districts start school well behind. Unless we know where students started (in the fall of kindergarten, not, as is now the case, in third grade), we have no idea how effective teaching actually is. When you compare scores of low-income students in urban communities with scores of the relatively few low-income students in the suburbs, the scores are very similar, but the state reports don’t make that clear. Indeed, we have had visitors to Bay State Reading Institute schools whose own children are in suburban systems tell us that the teaching they see in our urban districts is better than what their own kids get—even though test scores in their own district may be higher.

The current system unfairly demonizes thousands of urban district teachers who are working their hearts out to help the neediest kids and whose results are just as good as, if not better than, suburban schools when you adjust for where their students started. The so-called growth scores published by the state would appear to reflect these adjustments, but they don’t because the starting point is third grade and not kindergarten. The current system assumes that what happens between kindergarten and third grade doesn’t matter when in fact the opposite is true. When schools succeed, the improvement starts in kindergarten and then moves like a wave through succeeding grades.

Of course, good schools need testing to help teachers and the principal know which kids are falling behind and which interventions are working. A well-run district will have quick, formative assessments administered regularly during the year. The much shorter state assessment I’d like to see should be used to narrow the state’s attention to the relatively few districts where it needs to intervene. There’s little evidence that standardized tests have improved pedagogy, and plenty of evidence that they’ve reduced substantially the time available for instruction.

Many people who, like me, think we do too much testing link this to Common Core, which they oppose. But Common Core is a very good set of standards for what students should be learning; it is not the same as the tests. In English language arts, the emphasis is on having students read text for meaning, ask critical questions, look for evidence, pick out main ideas, and use material to articulate their own ideas. Having this new, well-thought-out, national standard has gotten teachers’ attention in a positive way. Having Massachusetts on the same page as other states means we now have access to curriculum materials designed specifically for the standards we’re teaching to.

I applaud the state for its support of Common Core; I doubt that people who’ve actually visited a Common Core classroom object to the teaching they saw there.

For me, the charter schools debate misses the most important question. The state currently has 4.2 percent of students in charters and that number would rise to 14 percent in a decade if the referendum passes. What matters more is what we do for the 86 percent of our students who will remain in regular district schools. But the charter debate sucks up all of the legislative and journalistic energy available to discuss school change, so no serious attention is given to investing in teachers and principals—even though there is strong evidence that doing so pays.

Massachusetts officials currently mandate that all districts use the same state-designed teacher evaluation system. While well-intentioned, the system has become time-consuming and bureaucratic. In a school where all students take regular formative assessments, where the principal meets regularly with teachers to discuss individual student data, and where the principal visits classrooms on a daily basis, the principal already knows which one or two teachers should be considered for dismissal, and which other teachers have plenty of potential but need a bit of help. Principals in schools like this tell me that the paperwork in the state system takes away from the time they have available to work with their teachers.

Meet the Author
For the great majority of districts, the one-size-fits all, paper-heavy state mandates and the overuse of testing are unlikely to improve the quality of instruction. This does not mean that what districts are now doing is good enough—far from it. If we want to use state policy as an engine for school improvement, we need a smarter strategy. We need a strategy designed to bring out the best in teachers and principals. Such a system would build on an understanding of the critical importance of strong educational leadership by superintendents and principals and would work to strengthen and inspire that leadership by carefully designed financial incentives and waivers from state mandates.

Edward Moscovitch is the executive director and cofounder of the Bay State Reading Institute. As a consultant to the Business Alliance for Education, he formerly helped draft and pass the 1993 Education Reform Law.