What is the difference between a sampling distribution and a sample?
To test the null hypothesis that the IV and DV are unrelated to each other, you use information from both your sample AND the sampling distribution of F. Here is what the process looks like: Step 1) You run a regression and you are given an F value as part of the output. This F value is based on your sample. Step 2) The F value calculated from your sample is compared to the sampling distribution of F. This sampling distribution is the distribution of F values we would expect to see if we drew thousands of samples and the null hypothesis were true. Stata tells you where the F value from your sample falls on the sampling distribution by giving you the “probability of F”. A low probability indicates that your sample F value is very unlikely to occur when the null hypothesis is true. Most researchers tend to reject the null hypothesis when the probability of F is below .05 or .01.
