A researcher wishes to compare how students at two different schools perform on a math test. He randomly selects 40 students form each of the schools and obtains their test scores. He pairs the first score from school A with the first score from school B, the second score from school A with the second score from school B and so on. He then performs a hypothesis test for matched pairs.
a.) is this approach valid? Why or why not
b.) If it is not a valid approach, how should the researcher have proceeded?