SAMPLE SIZE 1/12/16 Flashcards
1
Q
how do we need to design studies with small confidence limits?
A
- they have less variation in our data
- have a larger sample
2
Q
large sample size
A
relates to the power of the study
3
Q
What is power?
A
the probability of rejecting the null hypothesis when the null hypothesis is false
4
Q
Why do you calculate power?
A
BEFORE THE STUDY
- to be sure that your study is well designed
- justify your study to grant reviewers
AFTER
- to demonstrate that you were sufficiently well-powered to detect an association
- to estimate that if there is ana association, how small it must be for it to be missed
5
Q
Why do we estimate what sample size would give us small confidence intervals?
A
as complicating power can be complicated
6
Q
What is the odds of an event?
A
the probability an event occurs divided by the probability an event does not occur
7
Q
How do you calculate the odds ratio for 2 groups?
A
the odds of an event for exposed individuals divided by odds of an event for unexposed individuals
8
Q
What does it mean if the odds ratio = 1
A
there is no effect
