Question: A sports geek decided that the best way to predict the results of Big Ten Football games was to look at past records within the conference. He compiled data for the past 20 years, which was a total of about 160 games for each team.

He took random independent samples of size 12 of the scores of each team over the past 20 years and recorded the data. For the upcoming Indiana Hoosiers and Wisconsin Badgers game the sample data looked like this:

Indiana : n = 12 , sample scores: 36, 0, 14, 6, 12, 21, 17, 24, 6, 12, 21, 15.

Wisconsin: n = 12, sample scores: 14, 14, 12, 28, 32, 17, 12, 35, 12, 10, 28, 31.

The sports geek, regardless of whether you think his method of evaluating teams is valid, feels that Wisconsin will defeat Indiana by a touchdown (6 points). He is 90 % confident in his prediction system.

What inferences and conclusions can you draw about this problem? Does the data support the Sports Geek's system? You may use Excel or Minitab or process the data by hand. Also, complete both the Region Rejection Method and the p-value approach, and comment on whether or not the two methods are compatible.