Can anyone tell me how to calculate the sample size using power repeated command?
It seems that the power repeated command needs the alternative cell mean. What can we do if all we have is "mean decline"?
Hypertension example:
Suppose we are planning a longitudinal study to compare the mean change in SBP between a treated and a control group. It is projected, based on previous data, that the standard deviation of SBP at both baseline and follow-up is 15 mm Hg and that the correlation coefficient between repeated SBP values 1 year apart is approximately .70. How many participants do we need to study to have 80% power to detect a significant difference between groups using a two-sided test with α =.05 if the true mean decline in SBP over 1 year is 8 mm Hg for the treated group and 3 mm Hg for the control group?
The answer is 85 subjects in each group.
The solution is attached, thanks in advance!
It seems that the power repeated command needs the alternative cell mean. What can we do if all we have is "mean decline"?
Hypertension example:
Suppose we are planning a longitudinal study to compare the mean change in SBP between a treated and a control group. It is projected, based on previous data, that the standard deviation of SBP at both baseline and follow-up is 15 mm Hg and that the correlation coefficient between repeated SBP values 1 year apart is approximately .70. How many participants do we need to study to have 80% power to detect a significant difference between groups using a two-sided test with α =.05 if the true mean decline in SBP over 1 year is 8 mm Hg for the treated group and 3 mm Hg for the control group?
The answer is 85 subjects in each group.
The solution is attached, thanks in advance!
Comment