It is known that if people are asked to make an estimate of something, for example, “How tall is the University chapel?” the average guess of a group of people is more accurate than an individual’s guess. Vul and Pashler (2008) wondered if the same held for multiple guesses by the same person. They asked people to make guesses about known facts. For example,
“What percentage of the world’s airports are in the United States?” Three weeks later the researchers asked the same people the same questions and averaged each person’s responses over the two sessions. They asked whether this average was more accurate than the first guess by itself.
(a) What are the null and alternative hypotheses?
(b) What would be a Type I and Type II error in this case?
(c) Would you be inclined to use a one-tailed or a two-tailed test in this case?