By Scott Pearson
Ward 3 DC Board of Education Representative Ruth Wattenberg wrote an op-ed in The Washington Post calling for more and better data on DC’s public schools to help policymakers –particularly the State Board of Education – improve DC’s schools. Wattenberg specifically cited the need for deeper reporting on standardized test outcomes, school-by-school data on teacher retention, school-by-school reporting on time spent on testing, school-by-school reporting on curriculum changes over time, and better “school climate” data on each school.
Wattenberg is a thoughtful, and balanced education policymaker, part of a particularly strong recently elected cohort of Board of Education members. Her ideas should be taken seriously.
And it is in that spirit that I respectfully disagree with some – though not all of her points. My perspective comes from that of a leader in the charter school movement. Charter schools represent not only a new type of public school, they represent a new type of approach to school improvement.
The traditional approach to improving a public school system involves making changes centrally to the components, or “inputs” of a school system. To improve a large public school system, a leader might modify the curriculum, change the school staff or staffing model, reallocate resources, launch a school culture initiative, or put in place a new professional development program.
For public charter schools, the approach is different. Surely the leaders of each individual public charter school have the same toolkit to improve their school as the leader of a traditional school system. But at the city or state level, one wouldn’t dream of imposing such changes on all of the disparate public charter schools. Each school pursues its own path to improvement. At the system level, we improve quality not by dictating, or often even measuring the various steps public charter schools take to improve their schools. Instead we measure outcomes, like student performance and graduation rates, to determine which schools have been successful and which haven’t. Successful schools are allowed to grow. Unsuccessful schools are given time to turnaround, and ultimately are closed if they remain low performing. That approach has worked spectacularly well with DC’s public charter schools, whose students consistently perform better than the city average.
For this reason, charter advocates are instinctively skeptical about the idea that a state board, or any oversight body, can improve public charter schools by studying their curriculum, or staffing models, or schedules, and putting in place statewide policies to “improve” them. The diversity of school models makes such an approach virtually impossible. To take the issue of teacher attrition for instance, one public charter school might thrive on a high attrition model; another might lower attrition through changes to salary, and a third might already have low attrition. It’s the diversity of approaches that makes the whole charter sector stronger.
This is not to argue that we should be blind to information about what our schools are doing and how they are doing it. Education research is important and holds the promise to permit us to make great strides in the effectiveness of our schools. Some data on educational inputs could lead to beneficial statewide policies that are respectful of the diversity of public charter school approaches. For example we use data on the school facility expenditures to advocate for appropriate statewide expenditure. And parents need good data about schools to make appropriate choices for their children.
Finally, I certainly welcome any improvement in the quality of data on school outcomes, such as performance on standardized tests. That output data is vital to effective oversight of public charter schools.
But when collecting new data from schools, a healthy balancing of the costs of collecting this data needs to be balanced against an appropriately skeptical view of just how useful this data would be in a district made up of 63 independent, and highly individualistic public and public charter school districts.
The costs of data collection are high indeed.
The most obvious cost is staff time spent in collecting and reporting this data. Most DC schools already have a full-time data manager who does nothing but report data on important data such as attendance, truancy, discipline incidents and school performance. New burdens, such as tracking and reporting curriculum changes, collecting and reporting on the innumerable components of school “climate”, and any other potentially interesting information will require schools to divert another staff member, which could mean cutting the librarian, the psychologist, or the art teacher.
And make no mistake; the potential data to be collected is endless. The recent National Academy of Sciences report on DC’s education reform progress cited favorably a list of 189 separate data elements that DC should consider collecting, school-by-school – from data bandwidth per student to the percent of school trash that is recycled.
Sometimes the data collection impacts learning time as well. For example, the city’s desire to know how much children know about health topics means that schools now take hours of precious classroom time for health assessments.
There is a subtler, but perhaps greater threat to expanded data collection. In the hyper-choice DC education environment, schools are exquisitely sensitive to any data reported on them. The very collection of data can therefore influence a school’s decisions – and not always for the better. For example, suppose we started collecting and publicly reporting on teacher attrition. This could lead to schools feeling pressure to reduce attrition, which might be good – until a school hesitates on letting go a weak teacher. Who benefits then?
For these reasons, I am cautious about supporting new data requests. And when there are strong arguments for the data collection, I ask whether the data could be collected on a sample basis, to avoid burdening the schools, or whether the data can be anonymized, to avoid influencing school decisions.
DC already collects a lot of data. And perhaps there is more data that is likely to allow the city to make decisions that will improve educational outcomes. But let’s approach data collection with an appreciation of the cost of such collection, and hard questions about how useful such data will really be in a decentralized educational environment like Washington, DC.
Scott Pearson is the Executive Director of the DC Public Charter School Board.
PCSB Executive Director writes about how measuring outcomes, like student performance and graduation rates, has been successful for DC public charter schools.