Chapter 10 of 10All chapters
Chapter 10 of 10
Using it yourself
Beyond the laboratory.
Everyday experiments
Changing one thing at a time, keeping a record, and comparing against a baseline works for diet, sleep, workflow and code performance. Most personal experiments have a sample size of one and no control.
- Write down the prediction before you run the change.
- Beware the placebo effect on yourself; it does not stop working when you know about it.
Reading the news
Find the study behind the headline, check what was measured and in whom, and see whether the headline claims causation the design cannot support. That takes minutes and changes the conclusion often.