0% found this document useful (0 votes)
43 views25 pages

HED 296 CH 8 Sample Size and Power

This document discusses determining sample size for confidence intervals, hypothesis testing, and various study designs. It provides examples of calculating sample size needed for estimating a mean or proportion with a confidence interval of a specified width and power, and for hypothesis tests comparing means or proportions between groups with a given power to detect a specific effect size. Larger sample sizes are needed to achieve more precise estimates or greater power to detect smaller effects.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
43 views25 pages

HED 296 CH 8 Sample Size and Power

This document discusses determining sample size for confidence intervals, hypothesis testing, and various study designs. It provides examples of calculating sample size needed for estimating a mean or proportion with a confidence interval of a specified width and power, and for hypothesis tests comparing means or proportions between groups with a given power to detect a specific effect size. Larger sample sizes are needed to achieve more precise estimates or greater power to detect smaller effects.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 25

Chapter 8

Power and
Sample Size
Determination
Sample Size Determination

• Adequate sample size is needed to


ensure precision in analysis.
• Sample size determined based on type of
planned analysis.
– Confidence interval estimate
– Test of hypothesis
Determining Sample Size for
Confidence Interval Estimates
• Goal is to estimate an unknown
parameter using a confidence interval
estimate.
• Plan a study to sample individuals, collect
appropriate data and generate CI
estimate.
• How many individuals should we sample?
Determining Sample Size for
Confidence Interval Estimates
• Confidence intervals
point estimate ± margin of error
• Determine n to ensure small margin of error
(precision).
– Remember, width of CI is an indication of
precision.
• Must specify desired margin of error,
confidence level, and variability of parameter.
Example 8.1. Find n for One
Sample, Continuous Outcome
• Planning study to estimate mean systolic
blood pressure in children with
congenital heart disease.
• Want estimate within 5 units of true
mean
• Will use 95% confidence level; estimate
of standard deviation is 20.
Example 8.1. Find n for One
Sample, Continuous Outcome

2 2
 Zσ   1.96(20) 
n      61.5
 E   5 

• E is margin of error
• Need sample size of 62 children with
congenital heart disease
Example 8.3. Find n for One
Sample, Dichotomous Outcome
• Planning study to estimate proportion of
freshmen who currently smoke
• Want estimate within 5% of the true
proportion; will use 95% confidence level
Example 8.3. Find n for One
Sample, Dichotomous Outcome
2 2
 Z  1.96 
n  p(1  p)   0.5(1  0.5)   384.2
E  0.05 

• Formula requires estimate of proportion,


p. If unknown, use p = 0.5 to produce
largest n (most conservative).
• Need sample size of 385 freshmen
• Website – Sample Size Calculator
Example 8.5. Find n for Two Independent
Samples, Continuous Outcome
• Planning a study to assess the efficacy of
a new drug to raise HDL cholesterol
• Participants will be randomized to receive
either the new drug or placebo and
followed for 12 weeks.
• Goal is to estimate the difference in mean
HDL between groups (1 – 2).
Example 8.5. Find n for Two Independent
Samples, Continuous Outcome
• Want estimate of the difference to be no
more than 3 units
• We will use a 95% confidence interval.
• The estimate of the (common) standard
deviation in HDL is 17.1 (from previous
data).
• We expect 10% attrition over 12 weeks.
Example 8.5. Find n for Two Independent
Samples, Continuous Outcome

2 2
 Zσ   1.96(17.1) 
n i  2   2   249.6
 E   3 

• Need n1 = 250 and n2 = 250 with


complete outcome data
Example 8.5. Find n for Two Independent
Samples, Continuous Outcome
• Need n1 = 250 and n2 = 250 with complete
outcome data (at end of study)
• Need to account for 10% attrition
• How many subjects must be enrolled?
Example 8.5. Find n for Two Independent
Samples, Continuous Outcome
• Need n1 = 250 and n2 = 250 with complete
outcome data
• Account for 10% attrition:
90% Complete study
Participants
enrolled N = ? 10%
(500)
Lost to follow-up

N (to enroll)*(% retained) = 500


Need to enroll 500/0.90 = 556
Determining Sample Size for
Hypothesis Testing
= P(Type I error) = P(Reject H0|H0 true)
= P(Type II error) = P(Do not reject H0|H0 false)

Which is worse?

•Power = 1 – = P(Reject H0|H0 false)


•Power is probability that test correctly rejects the
null
 and Power
Determining Sample Size for
Hypothesis Testing
  and power are related to the sample
size, level of significance (), and the
effect size (difference in parameter of
interest under H0 versus H1).
– Power is higher with larger 
– Power is higher with larger effect size.
– Power is higher with larger sample size.
Effect size
Example 8.11.
Find n to Test H0: 0
• Planning study to test average health
expenditures on health
H0: = $3302 vs.
H1: ≠ $3302 at = 0.05
• Determine n to ensure 80% power to
detect a difference of $150 in mean
expenditures on (assume standard
deviation is $890, from previous study).
Example 8.11.
Find n to Test H0: 0
μ1 - μ 0 150
ES    0.17
σ 890
2
 Z1-α/2  Z1-β   1.96  0.84 
2

n        271.3
 ES   0.17 

• Need sample size of 272


Example 8.12.
Find n to Test H0: p = p0
• Planning study of patients with elevated
cholesterol
H0: p = 0.26 vs.
H1: p ≠ 0.26 at = 0.05
• Determine n to ensure 90% power to
detect a difference of 5% in the proportion
of patients with elevated LDL cholesterol.
Example 8.12.
Find n to Test H0: p = p0
p1 - p 0 0.05
ES    0.11
p 0 (1 - p 0 ) 0.26(1 - 0.26 )
2
 Z1-α/2  Z1-β   1.96  1.282 
2

n        868.6
 ES   0.11 

• Need sample size of 869


Example 8.14.
Find n1, n2 to Test H0: 2
• Planning study to evaluate efficacy of new
drug to lower systolic blood pressure
H0: 2 vs.
H1: ≠2 at = 0.05
• Determine n1 and n2 to ensure 80% power
to detect a difference of 5 units in means
(assume standard deviation is 19.0).
• Expect 10% attrition.
Example 8.14.
Find n1, n2 to Test H0: 2
μ1 - μ 2 5
ES    0.26
σ 19.0
2
 Z1-α/2  Z1-β 
2
 1.96  0.84 
n  2   2   232.0
 ES   0.26 

• Need samples of size n1 = 232 and n2 = 232


• Account for 10% attrition:
N (to enroll)*(% retained) = 464
Need to enroll 464/0.90 = 516
Example 8.18.
Find n1, n2 to Test H0: p1 = p2
• Planning study to find proportion of people
with hypertension using a new drug
H0: pp2 vs.
H1: p≠p2 at = 0.05
• Determine n1 and n2 to ensure 80% power
to detect a difference in proportions on the
order of 24% versus 30% in the new drug
and placebo treatments.
Example 8.18.
Find n1, n2 to Test H0: p1 = p2

p1 - p 2 0.06
ES    0.135
p(1 - p) 0.27(1- 0.27)
2
 Z1-α/2  Z1-β 
2
 1.96  0.84 
n  2   2   860.4
 ES   0.135 

• Need samples of size n1 = 861 and n2 = 861

You might also like