Sunday Review: A Decision Has to Be Made on Reopening Schools

Here is a digest of the best links I found each week for anyone who might be interested.

Pandemic: the reality we live in

US coronavirus new cases are breaking records each day. The virus is spreading faster in many states. We are starting to see the deaths ticking up. In the meantime, we are facing a tough decision to reopen schools. Schools do not have a simple on-off switch. To reopen schools will not just take a lot of money. Reopening schools without a careful plan is going to threaten lives of students, faculties and their families. But, there are also bigger risks of not returning to schools. Children will lose the precious learning time in their life. The loss will be suffered more by the poorest. There might be a bigger mental impact to kids before school age for being isolated too long.

“A decision has to be made”

But, it needs a careful plan relying on science-based evidence. Here is a detailed public health approach on how to reopen schools, and 8 basic steps.

Disruptions: how it might be different

We are living in a world, where people’s core beliefs are often better predictors of their policy views than any other factors (demographic, job, location, etc.) Those beliefs influence our own forecast of the future regarding global events, major policies, etc. A review of Tetlock’s Superforecasting summarizes the findings from the book’s study of a small group of “superforecasters” who constantly beat the predictions of others, including domain experts. It turns out the more confident people are, the worse the predictions turned out to be. “The most successful were those who were cautious, humble, numerate, actively open-minded, looked at many points of view, and updated their predictions (‘foxes’). ” It took a probability view and open-mind to better forecast the future. I like the question if policy will eventually accept this methodology as medicine did in the last century.

Insights: raise the bar for ourselves

Some of the great stories about how the biggest consumer apps got their first 1,000 users. The key takeaway is that you got to know who you wanna reach. Most popular strategies involve you to reach out to those people directly — in any possible way. The one I liked most is Pinterest’s founder changed all computers in the Apple store to Pinterest. That shows what it takes to start a company.

You normally don’t need to solve the cold start problem if you are building a new product on top of an existing brand. It’s relatively easy to drive some initial traffic to the new features from an existing product. However, to grow the features, it’s equally important to narrowly define the targeted users early on. There could be many creative ways to reach the users, only if you know who they are.

Leadership: the ideas for better collaboration

There is this theory of Managerialism and Civil Service to explain why the United States, and to some extent the rest of the Anglosphere, lags in infrastructure. I am not fully convinced that it can explain the difference between the situations in America and Asia, given managerialism is a common philosophy across Asia as well. However, it did point out some issues behind the popular managerialism philosophy that can be seen everywhere even in tech companies.

On one end, managerialism doesn’t mean it cannot build infrastructure, nor lack of long-term thinking. I have seen successful organizations of managerialism that deliver world-class infrastructures in the tech world (guess?). But, on the other end, managerialism fails at differentiation — be bold, audacious, and explore new territory. It’s hard for people without domain knowledge to be good at taking calculated risks in my opinion. That leads to the similar observation in civil service, where the leaders tend to start small, close to foreign knowledge, and fail to think over a long time horizon.

Economy: the science hidden among fallacies

Here is an empirical study of What Trait Affects Income the Most. There are some surprises, but a lot of the data are confirming many existing observations. It also has a quick tutorial on the methodology to dig into this type of questions.