Lecture 6C: Sample Size Calculation

Question 1

What is the primary objective of sample size calculation?

Answer 1

To ensure sufficient power to detect a significant difference or association

Question 2

Why is sample size calculation important in research planning?

Answer 2

To avoid missing clinically important differences and to meet requirements from grant agencies and scientific journals

Question 3

What factors does sample size calculation depend on?

Answer 3

• Study design (treatment, frequency, risk factor)
• Normality
• Frequency (incidence/prevalence)

Question 4

What is the Type I error usually set at in sample size calculations?

Answer 4

0.05

Question 5

What is the usual power (1-β) set at in sample size calculations?

Answer 5

0.80

Question 6

What is the formula for Standardized Effect Size (SES)?

Answer 6

SES = (mean1 - mean2) / pooled SD

Question 7

How is pooled standard deviation (SD) calculated?

Answer 7

Question 8

What factors can lead to a larger sample size requirement?

Answer 8

• Larger variability of pilot data
• Smaller effect size obtained from pilot study
• Increase in desired power

Question 9

What should be accounted for in sample size calculations regarding study participants?

Answer 9

Attrition or drop-outs (depends on length of study)

Question 10

Fill in the blank: Sample size calculation is usually done _______ you collect data.

Answer 10

[before]

Question 11

True or False: Larger effect sizes require smaller sample sizes to achieve statistical significance.

Answer 11

True

Question 12

How is sample sized calculated?

Answer 12

After SES is calculated, use table C2 on p.845 to find the number of subjects in
each group required

Example