Hey, so at (6:50) I think you made a mistake. To know the chance of a type 1 error based on 20 tests, you would not do 0.05*n (where n is the sample size), but you would do (1-0.05)^n. This would give you the percentage of chance of making a type 1 error. Also, at (9:59), to get the significance level needed, you would do 1- (1-0.05)^n based on the logic of the previous mistake. Except for that, it's a great video!
Hey, so at (6:50) I think you made a mistake. To know the chance of a type 1 error based on 20 tests, you would not do 0.05*n (where n is the sample size), but you would do (1-0.05)^n. This would give you the percentage of chance of making a type 1 error. Also, at (9:59), to get the significance level needed, you would do 1- (1-0.05)^n based on the logic of the previous mistake. Except for that, it's a great video!