Churches, schools, and other nonprofit institutions have held jelly bean counting contests where contestants compete for a prize by estimating the number of jelly beans in a large jar. There is evidence suggesting that averages of contestant estimates tend to be superior to specific individual estimates. Why might this be true?