Chapter 20 - Sample Size and Power

Question 1

What is sample size in stats? What is sample size in health sci?

Answer 1

In stats, the number of observations in a data set. In the health sciences, the sample size is usually the number of individual humans in the study population

Question 2

What is a Confidence interval (CI)?

Answer 2

A statistical estimate of the range of likely values of a parameter in a source population based on the value of that statistic in a study population

Question 3

What does a narrow CI indicate?

Answer 3

A narrow CI indicates more certainty about the value of the statistic than a wide CI

Question 4

Why are Large sample sizes better than Small sample sizes?

Answer 4

Large samples from a population produce narrower CI results for statistical measures, which leads to higher certainty). They also lead to more statistically significant results

Small samples leads to less certainty (represented by a wider CI)

Question 5

What is a sample size calculator?

Answer 5

A tool used to identify an appropriate estimate of participants to recruit for a quantitative study, more accurately called a sample-size estimator

Question 6

What is the difference between a Bias and an Error?

Answer 6

• Bias is a flaw in the way a study was designed or conducted that leads to an inaccurate result.
• An error happens randomly and is the difference between what you find in a study and what is real

Question 7

What are Type 1 vs. Type 2 errors?

Answer 7

• A Type 1 error is a False positive where a study population yields a statistically significant test result even though there actually is no significant difference that actually exists (the probability of there being one is represented by alpha)
• A Type 2 error is a False negative where a statistical test of data from a study population finds no significant result even though a significant difference actually exists (the probability of there being one is beta)

Question 8

What is power? What is it defined as?

Answer 8

• Power is the ability of a test to detect significant differences in a population when differences really do exist.
• Power is defined as 1 – β, so a 20% likelihood of a type 2 error (that is, β = 20%) corresponds to a power of 80%.

Question 9

Do studies with more participants have more or less power? Why?

Answer 9

More power. Because too few participants lack adequate power to detect meaningful differences or associations in source populations. It would not generate a significant difference