You want to conduct an experiment with your class to see if students prefer Coke or Pepsi.
a. Explain how you could do this, incorporating ideas of blinding and randomization, (i) with a completely randomized design and (ii) with a matched pairs design.
b. Which design would you prefer? Why?