Adding number of samples for t-statistics
Everytime that I have two different number of samples for both populations, do I have to add the number of samples to get the degrees of freedom? For example, I have to follow this procedure when calculating the other cases of confidence intervals seen previously? Or this is just done when the populations are independent and the variance is unknown but assumed as equal?
2 answers ( 0 marked as helpful)
Yes especially as the numbers in the samples are quite low. In the Part if we didnt know the population variance but had sample variance, then we would have used a T statistic of high degrees of freedom which is very close to the normal distribution.
Yes