1.6 Considering Sample Size Flashcards
When we talk about power, we are referring to the
probability of rejecting the null hypothesis
Studies may be underpowered if they have
too few participants
Studies may be overpowered if they have
too many participants
Estimating the right sample size increases the probability of _______________ and ensures that the study is both
detecting an effect; ethical and cost-effective
An a priori power analysis is conducted
before your study takes place
An a priori power analysis allows us to estimate the
sample size that we require to detect an effect based on our desired α level, the desired power level (1 - β), and the size of the effect
An a priori power analysis allows us to control for both
Type 1 error probability, Type 2 error probability and control the power of the test
A post-hoc power analysis is typically conducted
after your study takes place
In a post-hoc power analysis, we already know the
outcome of our analysis
A post-hoc power analysis allows us to use
the already known sample size (N), the set α level, and the specified effect size to return the power (1 - β) associated with the test
Unfortunately, during a post-hoc analysis, you only control for the
a and not the b
In psychology, we would generally set the desired alpha level at ________, and our beta level at _________.
05, i.e. α = 0.05; 0.20, i.e. β = 0.20
Ultimately, if we set our beta level at________ , the power we would get for our sample would be _______
0.20; 0.8
A power that is lower than 0.8, is typically considered to be:
too low for most areas of research