Probability and experimental design

Question 1

What are the benefits of using a larger sample size?

Answer 1

• Account for as much variation as possible
• Estimates are more accurate if we have more data
• Likely to be more representative of the population of interest

Question 2

What are the pros and cons of using a small sample size?

Answer 2

• Cheaper, easier, quicker to collect
• Estimates are likely to be less accurate
• May be less representative of the population

Question 3

In statistics, we need to make sure observations are…?

Answer 3

• Representative
• Independent
• Replicated
• Possible need of blind/double-blind trials
• Randomised control trials (RCTs are the gold standard)

Question 4

What is the purpose of a control group?

Answer 4

To quantify random variation without any treatment or intervention

Question 5

What is the difference between a blind and a double-blind experiment?

Answer 5

In blind experiments, the researchers know the treatment plan but the subjects don’t, whereas in double-blind trials, neither party knows

Question 6

What is the difference between a type I error and a type II error?

Answer 6

A type I error is a false positive whereas a type II error is a false negative

Question 7

What choices do we make based on our p value?

Answer 7

If p<0.05 we reject our null hypothesis
If p>0.05 we fail to reject the null hypothesis

Question 8

What does statistical power depend on?

Answer 8

• The size of the effect
• The variation in the data
• How large the sample size is