Power and Sample Size Flashcards

(12 cards)

1
Q

How large a sample do I need

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

precision analysis

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Two types of statistical error

A

Type 1 error (alpha)

Type 2 error (beta)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Statistical power

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Under Neyman-Pearson, alpha is

A

set by investigator

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Under Neyman-Pearson, beta

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Power analysis applies to

A

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)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

as alpha decreases

A

so will the power. It will decrease

But beta increased

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

given in a fixed sample size:

A

if alpha increases, beta decreases

if alpha decreases, beta increases

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

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

A

increases precision

allows a smaller effect to be detected

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Factors affecting statistical power

A
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
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

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

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly