I am doing a study on drug receptor occupancy using positron emission tomography and wish to determine the standard deviation associated with each measurement.
I am doing a study on drug receptor occupancy using positron emission tomography and wish to determine the standard deviation or SEM associated with each measurement.
We have several treatment groups with 4 subjects each, but use the average of both the untreated baseline and the treatment group for calculating receptor occupancy from the measured BPnd. How do you calculate the SD from these values?
It looks like you are talking about the estimate of the population variance using the means of the samples, i.e. variance between the groups.
If you have means of each of n groups X1, X2,.....Xn, then the mean of the means is
X=(X1+X2+..Xn)/n
The standard deviation of the sample means is
STD^2 means= [(X1-X)^2 +(X2-X)^2 +......]/n-1
Since the sample size in each group is 4, then the estimate of the population variance computed from the variability of the sample means (between the groups) is
STD^2 population (between groups) = 4* STD^2 means
Usually this is needed to form the F ratio=STD^2 between/STD^2 within to perform ANOVA.
There is no standard deviation unless you have at least two observations of something. Standard deviation of each measurement makes no sense. Redpectfully, David Booth
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling