The first few sentences says it all.
Quote:
My previous piece on this general topic was on the difference between people who take a science view of things and an engineering view of them. I explained how an engineering view, when faced with an unexpected failure of some sort, immediately stops whatever they're doing, reverts it if something was recently changed.
A "science" view holds that statistical power is enough.
Yep. It might look good on paper, but that doesn't always translate into how it manifests in practical application.
If a project doesn't work as planned, then something in your calculations is wrong. At this point you stop and figure out what that something is, as no progress can be made until you do.