Power and Sample Size

Question 1

How large a sample do I need

Answer 1

not too small bc if so it will be waste of money resources since you will not get anything out of that study

larger than necessary, study wastes resources

therefore must estimate power to determine sample size

Question 2

precision analysis

Answer 2

estimate how large a sample is needed to produce a given level of precision at a fixed confidence level

Question 3

Two types of statistical error

Answer 3

Type 1 error (alpha)

Type 2 error (beta)

Question 4

Statistical power

Answer 4

1-beta; probability that the null hypothesis will be rejected when it is false

Question 5

Under Neyman-Pearson, alpha is

Answer 5

set by investigator

Question 6

Under Neyman-Pearson, beta

Answer 6

is estimated and will be determined by such characteristics as sample size, trait variance and effect size

Question 7

Power analysis applies to

Answer 7

all major types of epi data:

1. dichotomous (binary)
2. multiple categories
3. continuous
4. time-to-event (in this case, power will be a function of the number of events)

Question 8

as alpha decreases

Answer 8

so will the power. It will decrease

But beta increased

Question 9

given in a fixed sample size:

Answer 9

if alpha increases, beta decreases

if alpha decreases, beta increases

Question 10

without changing the null hypothesis, the only way to increase power while using a smaller alpha level is to increase sample size

Answer 10

increases precision

allows a smaller effect to be detected

Question 11

Factors affecting statistical power

Answer 11

Z   = a standard normal  random variable
  μ1 = mean under HA
  μ0 = mean under H0
  σ  = population standard deviation
  Zα = critical value at significance level of α
  n  = sample size

Question 12

which will have more power? one sided or two sided?

Answer 12

with all else being equal, one sided test will have more power