4.Test yourself

Can you increase the power of an experiment by:

Using factorial designs?

Increasing sample size

False. Factorial designs help to increase 
generality but do not increase the power
True
False
True
False
True. Other things being equal a 
larger sample size increases power

Better randomisation ?

Avoiding bias  ?

False. Randomisation does not increase
power. And randomisation can not be 
"better". It is either done correctly or
 incorrectly
True
False
True
False
False. Power and bias are distinct 
and independent

Better statistical methods     ?

True
False

Better control of variability  ?

True
False
Home Top
True. Some statistical methods are more
powerful than others, e.g. non-parametric 
methods tend to be less powerful than 
parametric ones
True. Controlling variabil;ity reduces
"noise" so the experiment has a better
chance of detecting an effect