USING G*POWER SOFTWARE (Calculating Power) BIOSTATS

PART 3

Step-by-Step Guide to Assignment 2.3

Using the same data, WK1 (ATTACHED), you used in the Discussion Question from Week 1 and the related descriptive statistics from the Week 1 Assignment complete the following:

3. Perform a power analysis using G*Power using the actual sample size presented in the dataset for week 1 (the practice data set has 150) and an effect size of .30. What does this mean in terms of the study and the probability of experiencing a type 2 error?

Step 1. If you have not closed out of G*Power, click the Clear button in the Protocol of power analysis window after completing Problem 2.2. If you need to reopen G*Power, the following need to be set:

Test family should be “t test”

Statistical test = “Means: Difference between two independent means (two groups)”

Type of power analysis = “Post hoc: Compute achieved power – given α, sample size, and effect size”

Tail(s) = “Two”.

Step 3. Change Effect size d to 0.3. Keep α err prob set at 0.05.

Step 4. Referring back to the descriptive statistics in Week 1 practice problem 1.1, there were 150 participants; 41 with and 109 without hypertension.

Had hypertension
	Frequency	Percent	Valid Percent	Cumulative Percent
Valid	no	109	72.7	72.7	72.7
	yes	41	27.3	27.3	100.0
	Total	150	100.0	100.0

Step 5. Type 109 in Sample size group 1. Type 41 in Sample size group 2. Click Calculate.

G*Power output:

The Output shows there is only 37% power to detect a significant difference in the two groups with hypertension with a small effect size. With this power, there is a 63% probability of a Type 2 error (accepting or failing to reject a null-hypothesis that is false).