Field note · September 24, 2025
Course note — Designing an experiment that can succeed
Most experiments fail before launch, because they were never large enough to detect the effect anyone hoped for. The fix costs ten minutes and happens before you write code.
Revised Designing an experiment that can succeed this week. The change was small and the reason was not.
Most experiments fail before launch, because they were never large enough to detect the effect anyone hoped for. The fix costs ten minutes and happens before you write code.
The first draft explained the mechanism and stopped. What was missing was the failure: the thing that happens when you get it wrong, described concretely enough that someone recognises it later. A lesson that only teaches the correct version leaves you unable to spot the incorrect one, which is the situation you will actually be in.
It sits in the Experiments and causal claims module. The exercise is a thinking one rather than a query one — some of this material is about deciding what to compute, and that does not need a keyboard.
Ordering matters here more than in most courses. It follows Correlation, confounding, and Simpson's paradox and leads into Reading a result without fooling yourself, and reading it out of sequence mostly works but costs you the setup.
Free means free, and it also means we can rewrite it whenever it is wrong. No edition, no errata PDF, nothing to repurchase. The whole course is 36 lessons and the fixes land the day we find them.