Practice quiz: Why design matters

4 questions. Pick an answer to see if it is right. You can try as many times as you like. These are for practice: the final test has different questions.

1. 1.In Fisher's tea test, what is the chance of picking all four milk-first cups by pure guessing?

   1 in 21 in 81 in 701 in 1,000
2. 2.Why does a study need a control group?

   To double the dataTo show what would have happened anywayJournals require two groupsTo make the p-value smaller
3. 3.What does randomisation balance that careful matching cannot?

   NothingFactors you don't know about or can't measureOnly ageOnly the sample size
4. 4.In ML experiments, the 'control' is usually...

   the test seta baseline: same data and budget, without your changethe largest modelthe GPU