View on GitHub

Committee to Elect Piotr Mitros

For excellence in the Cambridge Public School District

Bringing CPSD business practices into the 21st century

While I’ve been impressed with most individuals in the district at all levels, the overall organizational dynamics connecting those individuals are broken. The school committee, as a body, is ineffective. Decisions are being made in all the wrong places. We have many programs with no real idea whether they work.

Many organizational processes are defunct. Have you tried going through the school lottery? Registering for afterschool? Simply moving to modern management techniques, improving transparency and communications, and moving towards digital processes can lower costs, save both parents and school employees time (giving more time to interact with students), and make everything work better.

I’ve been involved in three organizations since finishing my Ph.D. The first developed a medical imaging technology which has saved many lives. It sold for $200 million. The second developed a technology which has helped educate millions. It’s valued at over $1 billion. The third is edX, an MIT-Harvard education initiative, which has also positively impact millions of lives. In those roles, I’ve learned about managing modern, efficient organizations.

I’d like to bring that experience to bear to streamline and fix many of the organizational processes to have the district function like a modern, transparent, easy-to-engage-with organization. The key changes I’d like to see are:

Proper use of evaluation and assessment

The school has tremendous numbers of programs and initiatives with no real evaluation on return-on-investment or on effectiveness. More than a third of CPSD employees operate outside of the classroom, including many of the most highly paid. Are these positions effective? How effective are co-teachers? Do we need more of fewer? How much benefit do we see from the expensive busing scheme? We don’t really know. I would like to put in processes so we may evaluate what does and doesn’t work, and direct our funding and energy towards initiatives which demonstrably improve student outcomes. We need to know so we can invest more in things which work and less in things which do not.

Of course, evaluation and assessment has a long history of misuse in school systems for decades. Where assessment fails is when it is used in isolation or tied directly to incentives. There are many things we don’t know how to measure, and once assessment is tied to incentives, those tend to fall by the wayside. However, used correctly, with the understanding that much of evaluation is qualitative and needs to be combined with theoretical and substantive arguments, more knowledge is better.

However, used correctly, integrating different types of evidence, and not tied to stakes or accountability, students, teachers, and school systems do better with monitoring, feedback, and continuous assessment.

Transparency, communication, and collaboration

Along lines of this, I will work to improve transparency. As ground rules:

From there, the school should shift to processes where everything is open unless there is a compelling reason to do otherwise. Furthermore, the information should not only by open, but the school should take proactive steps to spread and share information about the school system, even to otherwise less engaged parents.

This is a quality issue. We have some of the best teachers in the world, and we need to find ways to spread best practices throughout the district. Match Education has used this to great effect. This is an equity issue. Parents without strong networks have a difficult time figuring out what’s going on. This is an empowerment issue too. People can’t be empowered to act if they don’t know what’s going on. This is also a coherency issue. While I wouldn’t propose extending control to afterschool programs, more mechanisms for voluntary coordination could help both sides.

As with any other kind of assessment, it’s important that such processes be judgment-free and geared towards feedback for improvement rather.

Fix the lottery

This is a little technical, but the CPSD school algorithm is broken. It leads to poor placement, and encourages gaming. Strong evidence of gaming in Boston lead Boston Public Schools to shift to a more reasonable algorithm. In Cambridge, we have similar evidence. Indeed, some programs (such as Chinese language immersion) are virtually impossible to be admitted into without gaming. There are matching algorithms which are better by most metrics; we should shift to any one of those.

For the nerds: The CPSD algorithm is similar to the Gale-Shapley algorithm. Each parent picks schools. The school district assigns each student a random number indicating what would be the school’s preference. The algorithm than optimizes matching. It does so apparently optimizing to the random number rather more so than to parent choices. Using a proper Gale-Shapley, and giving a different set of random numbers for each school, would lead to both better placements and eliminate any incentives to game.

For the non-nerds: In addition to a bad algorithm, we do a terrible job for kids with parents who don’t do the research. We can do much better with guiding all parents to reasonable placements.

For everyone: The concept behind the school lottery is based on a 1960-era goal of desegregation which was adversarial with most of the population. In 2017, most people want to be in integrated schools. While there are some effects we need to manage, a 2017 solution doesn’t need to be nearly as strict, harsh, or adversarial. Starting the school by prominently sticking labels on kids like “low-SES” and “high-SES” on kids leads to negative and segregatory effects.

Reduce regulation

The school district has a tremendous number of very well-thought-out policies. They individually make sense, but together, they’re disempowering to teachers and prevent continued improvement and innovation. I would like to convert as many of these as possible from regulations into guidelines.

Policy is the opposite of personalization. Policies look at students as aggregates, while teachers can look at them as individuals. Many of the school policies, especially with regards to curriculum mandates, are in direct opposition to the district’s failing push for differentiated instruction in all classrooms, or for meeting the needs of learners from different backgrounds.

Engage with researchers

We have some of the best educational research in the world right here in Cambridge.

Why aren’t we making more use of it?

Part of the answer is that CPSD is difficult for researchers to engage with. We have no established processes for doing so, and principals and teachers must wing it every time. I don’t want to dictate a process, but I’d like to provide a default set of guidelines so that busy principals and teachers have an easy way to understand how to respond to researcher requests, and I’d like to let researchers know we’re open for business with an understandable set of guidelines.

Working with Harvard, MIT, Tufts, and others is one of the best ways to understand what happens in our district, as well as to bring in new ideas for how to improve it.

Engage teachers in the community

Did you know many CPSD teachers can’t send their own kids here? The wages from the schools no longer support the cost-of-living in Cambridge, and the school does not allow employees to send their own kids to the school. Good companies have employees use their own products. Employees can’t make informed decisions otherwise, and teachers, likewise, should be able to see how schools impact their own kids. That also brings teachers into the community, give teachers a heck of a lot more time (whether for work or family), and is just basic decency as an employer.

School committee politics

The school committee processes don’t work well. Visit a meeting or read any Cambridge Day article. I’d like to shift away from sparring, politics, and process to getting real work done.

Organizational roles

The role of the school committee ought to be the same as of a corporate board – high level vision and strategy. The role of the school administration ought to be to work with the school committee on defining that and to execute on it. Most of the innovation should come from the bottom up. The school committee should not micromanage (and neither should higher levels of the administration).

Go digital

Parents should be able to register for the school, afterschool, and other activities online, be notified of school events without paper, and monitor their childrens’ progress. In 2017, virtually everyone, including poorer individuals, have at least some internet access. I don’t disagree with allowing people to manage processes without it, but for the most part, we can save huge amounts of peoples’ time and the district’s money by going digital whereever possible.