Dear all
I have a question and I would be very grateful for some advice.
Is there any relationship between the standard deviations of two
samples and the standard error of the difference between two samples?
Say we have 40 individuals of the same age, 20 individuals are male
and 20 individuals are female. For each individuals we measure
height. I would now run a simple linear regression with height as the
response variable and sex as the dummy predictor variable (eventually
adjusted from some other variables). What I will get is an estimate
of the coefficient, i.e. the mean difference in height between males
and females and a standard error of this difference, which allows me
to calculate a confidence interval for the mean difference. Is there
any way to derive from the results of a linear regression what the
standard deviation of height in each subset of individuals (i.e.,
males and females separately) would be?
Thanks for any hints.
Pedram
|