It seems that you have missed the necessary options for us to answer this question, so I had to look for it. Anyway, here is the answer. Since it has been reported that half of americans show road rage and for this claim to have strong external validity, the one that would have been the best sample for the researcher to have used is this: a cross-section of American drivers representing men and women; people from the city,suburbs, and farms; of all ages.